Calibration curve generation device, calibration curve generation method and calibration curve generation program

Abstract

To provide a calibration curve generation device, a calibration curve generation method and a calibration curve generation program, capable of reducing an erroneous determination of an inspection result by an inspection device.SOLUTION: A calibration curve generation device generates a calibration curve to be referred to by an inspection device configured to inspect an article according to a spectroscopic spectrum of the article measured by a spectrophotometer. The calibration curve generation device includes a calibration curve calculation unit configured to calculate a calibration curve from a plurality of absorption spectra respectively obtained by the spectrophotometer for a plurality of articles, and known true values for the plurality of articles. The calibration curve calculation unit classifies component data composing each absorption spectrum of the plurality of absorption spectra into wavelength data D1 to D3 by wavelength, and removes a piece of the component data of a wavelength at which a prescribed condition is not met by wavelength data from the plurality of absorption spectra to calculate a calibration curve.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a calibration curve generating device, a calibration curve generating method, and a calibration curve generating program. [Background technology]

[0002] Patent Document 1 discloses an inspection device that detects whether the components of a molded product are appropriate by directing the transmitted light emitted from a light source that has passed through the molded product into a sensor as signal light and analyzing the signal light. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-112199 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the inspection device described in Patent Document 1 has a problem in that it analyzes signal light including unnecessary wavelengths, which can lead to erroneous judgment of the inspection results.

[0005] The present invention has been made in consideration of the above-mentioned circumstances, and aims to provide a calibration curve generating device, a calibration curve generating method, and a calibration curve generating program that can reduce the likelihood of the inspection device making an erroneous judgment on the inspection results. [Means for solving the problem]

[0006] The calibration curve generating device according to the present invention is a calibration curve generating device (2) that generates a calibration curve to be referenced by an inspection device (11) that inspects an article based on the spectroscopic spectrum of the article measured by a spectroscopic measuring device (10), and includes a calibration curve calculation unit (22) that calculates the calibration curve from a plurality of absorption spectra obtained by the spectroscopic measuring device for a plurality of articles and known true values ​​for the plurality of articles, and the calibration curve calculation unit is configured to classify component data constituting each absorption spectrum of the plurality of absorption spectra into wavelength data for each wavelength, and calculate the calibration curve by excluding component data of wavelengths for which the wavelength data does not satisfy predetermined conditions from the plurality of absorption spectra.

[0007] With this configuration, the calibration curve generating device of the present invention generates a calibration curve excluding component data whose wavelength data does not satisfy specified conditions, thereby reducing the chance of the inspection device making an erroneous judgment on the inspection results.

[0008] In the calibration curve generating device according to the present invention, the calibration curve calculation unit may calculate a correlation index that represents a correlation between the wavelength data and true value data consisting of true values ​​of the plurality of articles, and determine whether or not the wavelength data satisfies the predetermined condition according to the correlation index.

[0009] With this configuration, the calibration curve generating device according to the present invention determines whether the wavelength data satisfies a predetermined condition based on the correlation index between the wavelength data and the true value data, and can generate a calibration curve excluding component data whose wavelength data does not satisfy the predetermined condition.

[0010] Furthermore, in the calibration curve generating device according to the present invention, the calibration curve calculation unit may be configured to determine that the wavelength data satisfies the predetermined condition if the correlation index is equal to or greater than a predetermined threshold, and to determine that the wavelength data does not satisfy the predetermined condition if the correlation index is less than the threshold.

[0011] With this configuration, the calibration curve generating device according to the present invention can determine whether the wavelength data satisfies a predetermined condition according to the correlation index between the wavelength data and the true value data.

[0012] A calibration curve generation method according to the present invention is a calibration curve generation method for generating a calibration curve to be referenced by an inspection device (11) that inspects an article based on the spectroscopic spectrum of the article measured by a spectroscopic measurement device (10), and includes a calibration curve calculation step of calculating the calibration curve from a plurality of absorption spectra obtained by the spectroscopic measurement device for a plurality of articles and known true values ​​for the plurality of articles, wherein the calibration curve calculation step classifies component data constituting each absorption spectrum of the plurality of absorption spectra into wavelength data for each wavelength, and calculates the calibration curve by excluding component data for wavelengths for which the wavelength data does not satisfy predetermined conditions from the plurality of absorption spectra.

[0013] In this way, the calibration curve generation method of the present invention generates a calibration curve by excluding component data whose wavelength data does not satisfy the specified conditions, thereby reducing the chance of the inspection device making an erroneous judgment on the inspection results.

[0014] The calibration curve generation program according to the present invention is a calibration curve generation program that causes a computer to generate a calibration curve to be referenced by an inspection device (11) that inspects an article based on the spectroscopic spectrum of the article measured by a spectroscopic measurement device (10), and includes a calibration curve calculation step that calculates the calibration curve from a plurality of absorption spectra obtained by the spectroscopic measurement device for a plurality of articles and known true values ​​for the plurality of articles, and the calibration curve calculation step classifies component data constituting each absorption spectrum of the plurality of absorption spectra into wavelength data for each wavelength, and calculates the calibration curve by excluding component data for wavelengths for which the wavelength data does not satisfy predetermined conditions from the plurality of absorption spectra.

[0015] In this way, the calibration curve generation program of the present invention generates a calibration curve excluding component data whose wavelength data does not satisfy the specified conditions, thereby reducing the chance of the inspection device making an erroneous judgment on the inspection results. [Effects of the Invention]

[0016] According to the present invention, it is possible to provide a calibration curve generating device, a calibration curve generating method, and a calibration curve generating program that can reduce the possibility of an inspection device making an erroneous judgment on an inspection result. [Brief explanation of the drawings]

[0017] [Figure 1] FIG. 1 is a functional block diagram of an article inspection device equipped with a calibration curve generating device according to one embodiment of the present invention. [Figure 2] FIG. 2 is a conceptual diagram for explaining the operation of a calibration curve calculation unit that constitutes the calibration curve generating device according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] An article inspection system 100 according to an embodiment of the present invention includes an article inspection device 1 and a calibration curve generating device 2. As shown in FIG.

[0019] When an item to be inspected, which is being transported individually along a transport path by a transport unit, reaches a predetermined inspection position, the item inspection device 1 irradiates light onto the item, which is in a fixed position at the predetermined inspection position, and inspects the quality of the item based on the spectral spectrum of the transmitted light that passes through the item upon irradiation with this light (also called irradiated light).

[0020] The items to be inspected include unpackaged items with an outer diameter of several to several tens of mm that can be transported individually, bite-sized items, as well as items and molded products of a predetermined shape manufactured using existing manufacturing equipment or manufacturing equipment without inspection capabilities, and especially items that do not change shape during transportation.

[0021] Examples of such articles include pharmaceutical preparations such as tablets, capsules, lozenges, and drops, as well as candy and chocolate. The following description will be given taking as an example an article to be inspected a tablet W that is circular in plan view, has a height (thickness) smaller than its diameter, and is roughly cylindrical in side view. Note that the article to be inspected is not limited to a circular shape in plan view, and articles of various shapes such as an oval shape or a polygonal shape can also be used.

[0022] Examples of the conveying unit include a conveying belt, a conveying disk, a conveying chute, and the like, which are configured to align and convey articles individually.

[0023] The article inspection device 1 according to this embodiment includes a spectroscopic measurement device 10 and an inspection device 11.

[0024] The spectroscopic measurement device 10 includes a light source unit and a light detection unit. The spectroscopic measurement device 10 irradiates a tablet W to be measured with broadband light (visible light, near-infrared to terahertz light (terahertz waves)), and measures the spectrum of the light transmitted through the tablet W in response to the irradiation of this light.

[0025] In this embodiment, the inspection device 11 is composed of a computer unit equipped with a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), a storage device such as a hard disk drive, and a communication port.

[0026] The ROM and storage device of this computer unit store a program for causing the computer device to function as the inspection device 11. That is, the CPU executes the program stored in the ROM and storage device using the RAM as a working area, causing the computer unit to function as the inspection device 11 in this embodiment.

[0027] The inspection device 11 calculates the absorbance A of the tablet W at each wavelength λ from the light intensity Ii at the wavelength λ of the reference spectrum stored in the storage device and the light intensity I at the wavelength λ of the spectrum of the tablet W measured by the spectrometer 10, as follows: A=-log 10 The absorption spectrum of the tablet W is calculated by performing calculation according to (I / Ii). In this embodiment, the spectrum measured by the spectroscopic measurement device 10 in the absence of a measurement object is used as the reference spectrum.

[0028] The inspection device 11 performs preprocessing for a predetermined wavelength range of the absorption spectrum of the tablet W. The predetermined wavelength range is received from the calibration curve generating device 2 connected via a communication port and stored in the storage device of the inspection device 11.

[0029] The preprocessing performed by the inspection device 11 is the same as the preprocessing performed by the preprocessing unit 20 of the calibration curve generating device 2 described later, and therefore will not be described again. When the preprocessing of the absorption spectrum of the tablet W is completed, the inspection device 11 refers to the calibration curve stored in the storage device and calculates a measurement value corresponding to the absorption spectrum of the tablet W.

[0030] In this embodiment, the measured value represents the content of the test component contained in the tablet W. The calibration curve is received from the calibration curve generating device 2 connected via a communication port and stored in the storage device of the inspection device 11.

[0031] The inspection device 11 judges whether the quality of the tablets W is good or bad based on the measured values. The inspection device 11 outputs a sorting signal based on the quality or bad result of the judgment to a sorting device (not shown) that sorts the tablets W into normal products and defective products. The sorting device sorts the tablets W into normal products and defective products based on the sorting signal.

[0032] The calibration curve generating device 2 generates a calibration curve to be referenced by the inspection device 11 in order to inspect the tablet W measured by the spectroscopic measurement device 10. In this embodiment, the calibration curve generating device 2 is configured by a computer device including a CPU, RAM, ROM, a storage device such as a hard disk drive, a communication port, a display device, and an input device.

[0033] The ROM and storage device of this computer device store a program for causing the computer device to function as the calibration curve generating device 2. That is, the CPU executes the program stored in the ROM and storage device using the RAM as a working area, causing the computer device to function as the calibration curve generating device 2 in this embodiment.

[0034] The display device is, for example, a liquid crystal display device. The input device is, for example, a keyboard device or a pointing device. The input device may be, for example, a touch pad integrated with the display device. A cable for communicating with the inspection device 11 is connected to the communication port.

[0035] The storage device of the calibration curve generating device 2 stores the absorption spectra of a plurality of tablets whose true values ​​are known in association with the true values. The true values ​​of the plurality of tablets are measured in advance by an apparatus capable of high-precision measurement, such as a high-performance liquid chromatography apparatus. The absorption spectra of the plurality of tablets are measured in advance by the article inspection device 1.

[0036] In this embodiment, the absorption spectra of a plurality of tablets are grouped according to true values ​​and stored in the storage device of the calibration curve generating device 2. For example, the absorption spectra of 80 tablets with true values ​​near 9 are assigned to the first group. The absorption spectra of 80 tablets with true values ​​near 6 are assigned to the second group. The absorption spectra of 80 tablets with true values ​​near 2 are assigned to the third group.

[0037] The calibration curve generating device 2 includes a preprocessing unit 20, a classification unit 21, a calibration curve calculation unit 22, and a calibration curve evaluation unit .

[0038] The preprocessing unit 20 performs predetermined preprocessing on the absorption spectra stored in the storage device to convert them into a data format suitable for multivariate analysis, which will be described later. The predetermined preprocessing includes at least one of the four arithmetic operations, differentiation, integration, and normalization.

[0039] The classification unit 21 classifies the preprocessed absorption spectra into teacher data Tr and test data Te. In this embodiment, the classification unit 21 classifies the preprocessed absorption spectra equally into teacher data Tr and test data Te.

[0040] The calibration curve calculation unit 22 calculates a calibration curve that represents the correlation between the teacher data Tr and the true value by performing multivariate analysis on the teacher data Tr and the true value associated with the teacher data Tr.

[0041] The calibration curve calculation unit 22 performs multivariate analysis using, for example, simple regression analysis, multiple regression analysis, quantification type 1, quantification type 2, quantification type 3, discriminant analysis, logistic regression analysis, principal component analysis, partial least squares regression (PLS regression), factor analysis, cluster analysis, correspondence analysis, multidimensional scaling, conjoint analysis, support vector machine, decision tree, random forest, naive Bayes, neural network, deep learning, etc. In this embodiment, the calibration curve calculation unit 22 performs multivariate analysis using PLS regression.

[0042] 2, the calibration curve calculation unit 22 classifies the component data constituting each teaching data Tr into wavelength data for each wavelength. Note that while Fig. 2 shows an example in which the calibration curve calculation unit 22 classifies the component data constituting each teaching data Tr into wavelength data D1 to D3 for each of three wavelengths, this does not limit the number of wavelength data to be classified by the calibration curve calculation unit 22.

[0043] The calibration curve calculation unit 22 calculates a calibration curve by excluding wavelength component data whose wavelength data does not satisfy a predetermined condition from the plurality of teacher data Tr. Specifically, the calibration curve calculation unit 22 calculates a correlation index that represents the correlation between the wavelength data and true value data consisting of true values, and determines whether the wavelength data satisfies the predetermined condition based on the correlation index. In this embodiment, the calibration curve calculation unit 22 calculates a correlation coefficient as the correlation index between the wavelength data and the true value data.

[0044] If the correlation coefficient is equal to or greater than a predetermined threshold value TH, the calibration curve calculation unit 22 determines that the wavelength data meets the predetermined conditions, and if the correlation coefficient is less than the threshold value TH, it determines that the wavelength data does not meet the predetermined conditions.

[0045] Hereinafter, the operation of the calibration curve calculation unit 22 will be specifically described with reference to FIG.

[0046] 2 shows an example in which the training data Tr for tablet A, whose true value is 9, is composed of component data of [400, 300, 200] for each wavelength, the training data Tr for tablet B, whose true value is 6, is composed of component data of [500, 150, 210] for each wavelength, and the training data Tr for tablet C, whose true value is 2, is composed of component data of [20, 120, 180] for each wavelength. Note that, although FIG. 2 shows three training data Tr to make the invention easier to understand, the number of training data Tr is not limited.

[0047] The calibration curve calculation unit 22 classifies the component data of [400, 500, 20] into wavelength data D1, the component data of [300, 150, 120] into wavelength data D2, and the component data of [200, 210, 180] into wavelength data D3.

[0048] The calibration curve calculation unit 22 calculates the correlation coefficient between the true value data consisting of the true value [9, 6, 2] and each of the wavelength data D1 to D3. In the example shown in Fig. 2, the correlation coefficient between the true value data and the wavelength data D1 is 0.8, the correlation coefficient between the true value data and the wavelength data D2 is 0.9, and the correlation coefficient between the true value data and the wavelength data D3 is 0.7.

[0049] If the correlation coefficient is greater than or equal to a threshold value TH (0.8 in this embodiment), the calibration curve calculation unit 22 determines that the wavelength data meets the predetermined conditions, and if the correlation coefficient is less than the threshold value TH, it determines that the wavelength data does not meet the predetermined conditions.

[0050] 2, it is determined that wavelength data D1 and D2 satisfy the predetermined conditions, and wavelength data D3 does not satisfy the predetermined conditions. Therefore, the calibration curve calculation unit 22 calculates a calibration curve based on the usage data obtained by excluding, from each teacher data Tr, the component data classified as wavelength data D3 whose wavelength data does not satisfy the predetermined conditions. The calibration curve calculated by the calibration curve calculation unit 22 is transmitted to the inspection device 11 via a communication port in response to, for example, an operation of the input device, and is stored in the storage device of the inspection device 11.

[0051] The calibration curve evaluation unit 23 calculates an evaluation value of the calibration curve calculated by the calibration curve calculation unit 22 using the test data Te and the true value associated with the test data Te, and evaluates the accuracy of the calibration curve based on the evaluation value.

[0052] In this embodiment, the calibration curve evaluation unit 23 calculates an evaluation value of the calibration curve based on at least one index such as the mean square error, the root mean square error, the mean absolute error, the average error, the prediction standard error, the coefficient of determination, and the coefficient of determination corrected for degrees of freedom, etc. The calibration curve evaluation unit 23 displays the calculated evaluation value on, for example, a display device.

[0053] As described above, the calibration curve generating device 2 according to this embodiment generates a calibration curve excluding component data whose wavelength data does not satisfy predetermined conditions, thereby reducing the likelihood of the inspection device 11 making an erroneous judgment on the inspection results.

[0054] Furthermore, the calibration curve generating device 2 according to this embodiment determines whether the wavelength data satisfies a predetermined condition based on the correlation index between the wavelength data and the true value data, and can generate a calibration curve excluding component data whose wavelength data does not satisfy the predetermined condition.

[0055] Furthermore, the calibration curve generating device 2 according to this embodiment determines that the wavelength data meets the predetermined conditions if the correlation index is equal to or greater than a predetermined threshold value TH, and determines that the wavelength data does not meet the predetermined conditions if the correlation index is less than the threshold value TH, and is therefore able to determine whether the wavelength data meets the predetermined conditions according to the correlation index between the wavelength data and the true value data.

[0056] In the present embodiment, an example has been described in which the calibration curve calculation unit 22 calculates a correlation coefficient as a correlation index between wavelength data and true value data. However, the calibration curve calculation unit 22 may calculate any other value as a correlation index as long as the value represents the correlation between wavelength data and true value data.

[0057] For example, the calibration curve calculation unit 22 may calculate, as a correlation index, a value that indicates the tendency for the wavelength data and the true value data to be proportional to each other, or may calculate, as a correlation index, a value that indicates the tendency for the wavelength data and the true value data to be proportional to each other by converting them using a relational expression.

[0058] While embodiments of the present invention have been disclosed above, it will be apparent to those skilled in the art that modifications may be made without departing from the scope of the present invention. All such modifications and equivalents are intended to be encompassed by the following claims. [Explanation of symbols]

[0059] 2. Calibration curve generator 10 Spectrometer 11 Inspection equipment 22 Calibration curve calculation section

Claims

1. A calibration curve generating device (2) that generates a calibration curve to be referenced by an inspection device (11) that inspects an article based on a spectroscopic spectrum of the article measured by a spectroscopic measurement device (10), comprising: a calibration curve calculation unit (22) that calculates the calibration curve from a plurality of absorption spectra obtained by the spectroscopic measurement device for a plurality of articles and known true values ​​for the plurality of articles; The calibration curve calculation unit classifying component data constituting each of the plurality of absorption spectra into wavelength data for each wavelength; a calibration curve generating device that calculates the calibration curve by excluding, from the plurality of absorption spectra, component data of wavelengths for which the wavelength data does not satisfy a predetermined condition;

2. The calibration curve calculation unit calculating a correlation index representing a correlation between the wavelength data and true value data consisting of true values ​​of the plurality of articles; The calibration curve generating device according to claim 1 , wherein whether or not the wavelength data satisfies the predetermined condition is determined based on the correlation index.

3. The calibration curve calculation unit 3. The calibration curve generating device according to claim 2, wherein if the correlation index is equal to or greater than a predetermined threshold, it is determined that the wavelength data satisfies the predetermined condition, and if the correlation index is less than the threshold, it is determined that the wavelength data does not satisfy the predetermined condition.

4. A calibration curve generating method for generating a calibration curve to be referenced by an inspection device (11) that inspects an object based on a spectroscopic spectrum of the object measured by a spectroscopic measurement device (10), comprising: a calibration curve calculation step of calculating the calibration curve from a plurality of absorption spectra obtained by the spectroscopic measurement device for a plurality of articles and known true values ​​for the plurality of articles, The calibration curve calculation step classifying component data constituting each of the plurality of absorption spectra into wavelength data for each wavelength; A calibration curve generating method for calculating the calibration curve by excluding component data of wavelengths for which the wavelength data does not satisfy a predetermined condition from the plurality of absorption spectra.

5. A calibration curve generation program that causes a computer to generate a calibration curve to be referenced by an inspection device (11) that inspects an item based on a spectroscopic spectrum of the item measured by a spectroscopic measurement device (10), comprising: a calibration curve calculation step of calculating the calibration curve from a plurality of absorption spectra obtained by the spectroscopic measurement device for a plurality of articles and known true values ​​for the plurality of articles, The calibration curve calculation step classifying component data constituting each of the plurality of absorption spectra into wavelength data for each wavelength; a calibration curve generating program for calculating the calibration curve by excluding, from the plurality of absorption spectra, component data of wavelengths for which the wavelength data does not satisfy a predetermined condition;

Images