Product inspection method and product inspection device

Abstract

The present invention provides a new technology for high-speed and high-reliability good or bad judgment through optical measurement. Superlattice light having a continuous spectrum at least in the range of 1100 to 1300 nm is emitted from a pulse light source (1), and is pulse-stretched by an elongation element (2) in a manner such that the relationship between the wavelength in a pulse and the elapsed time becomes 1:1, and irradiated to a product (P). The light passing through the product (P) is received by a photoreceiver (3), and the output data is input to a judgment unit (7). The good or bad judgment program (43) of the judgment unit (7) calculates the absorption spectrum based on the output data, quantifies it using chemometrics, and compares it with a reference value to judge whether it is good or bad. Products (P) judged to be defective are excluded by an exclusion mechanism (8).

Description

Technical Field

[0001] The present invention relates to a technique for inspecting products for quality by optical measurement of various products. Background Art

[0002] A technique of spectroscopic analysis in which an object is irradiated with light and the transmitted light, reflected light, etc. from the object are spectroscopically analyzed to measure a spectrum is one of the representative techniques as a method for material analysis. This method is also applied to the judgment of product quality.

[0003] Typically, as in HPLC (high performance liquid chromatography), a part of the product is taken out, dissolved in a solution as needed, and its absorption spectrum, etc. is measured, and thus the quality of the product is judged. The absorption spectrum, etc. in the case of a qualified product is investigated in advance, and the quality of the product is judged by comparing it therewith.

[0004] Prior Art Documents

[0005] Patent Documents

[0006] Patent Document 1: Japanese Patent Laid-Open No. 06-034622

[0007] Non-Patent Documents

[0008] Non-Patent Document 1: Edited by Yukihiro Ozaki, published by KODANSHA LTD., "Near-Infrared Spectroscopy", pages 59-75

[0009] Summary of the Invention

[0010] Technical Problem to be Solved by the Invention

[0011] Light characteristics such as absorption spectra and reflection spectra faithfully reflect the contained components and their amounts, and component identification and quantification can be performed with high precision. Therefore, in product inspection, good judgment can also be made with high reliability.

[0012] However, in the conventional method such as HPLC, there is a drawback that the measurement or its preparation is very time-consuming. Therefore, it remains at the level of periodically taking out a part of the product and confirming the quality. For products that require particularly high reliability such as pharmaceuticals, full inspection is preferably performed in most cases, but it is unrealistic to perform full inspection by optical measurement in the conventional method.

[0013] For each completed product, instead of dissolving a part in a solution, light is directly irradiated, the transmitted light from the product is spectroscopically measured, and thus an absorption spectrum is obtained, and good judgment is made therefrom. However, considering that in the conventional spectroscopic measurement using a diffraction grating, it is necessary to change the posture of the diffraction grating (wavelength scanning) in accordance with the measurement band, and high-speed measurement assuming full inspection cannot be achieved.

[0014] In addition, in the case of absorbing a product with a high absorption rate (transmittance less than about 1%), the transmitted light is weak, so problems in measurement accuracy are expected to occur. In spectroscopic measurement using a diffraction grating, in order to sufficiently improve the SN ratio of the measurement or perform highly sensitive measurement, it is necessary to slow down the scanning or perform scanning multiple times to increase the total amount of light (light quantity) incident on the light receiver. This is one of the main reasons why high-speed measurement is difficult, but in the case where the light to be measured is further weakened, this problem becomes significant. That is, if high-speed priority is given, due to the influence of the SN ratio, the inspection accuracy is significantly reduced.

[0015] In addition, if a multi-channel spectroscopic measurement device using an area sensor in which a plurality of photoelectric conversion elements are arranged in a row is used, scanning of the diffraction grating is not required. However, in order to perform high-SN ratio or high-sensitivity analysis, it is necessary to increase the light quantity, and the problem of inability to perform high-speed analysis cannot be solved.

[0016] The present invention is completed to solve such problems of the prior art, and an object thereof is to provide a new technique for performing high-speed and highly reliable determination of good or bad through optical measurement.

[0017] Means for Solving the Problem

[0018] To solve the above problems, a product inspection method of the present invention is a product inspection method for determining whether a product is good or bad by performing optical measurement, and includes:

[0019] An emission step of emitting broadband pulsed light from a light source;

[0020] An elongation step of elongating the pulse width of the emitted pulsed light by an elongation element so that the relationship between the wavelength and the elapsed time in one pulse becomes 1:1;

[0021] An irradiation step of irradiating the elongated pulsed light onto the product;

[0022] A light receiving step of receiving light from the product irradiated with the elongated pulsed light by a light receiver; and

[0023] A determination step of processing the output data from the light receiver to determine whether the product is good or bad.

[0024] In addition, to solve the above problems, in this product inspection method, the determination step is a step of calculating the content ratio or content of a specific component of the product from the output data, comparing the calculated content ratio or content with a reference value, and thereby determining whether the product is good or bad.

[0025] In addition, to solve the above problems, in this product inspection method, the determination step is a step of comparing the value at a specified time within one pulse in the output data from the light receiver with a reference value without calculating the optical characteristics, thereby determining whether it is good or not.

[0026] In addition, to solve the above problems, in this product inspection method, the determination step is a step of comparing the optical characteristics of a specified wavelength calculated based on the value at a specified time within one pulse in the output data from the light receiver or the value at the specified time with a reference value, thereby determining whether it is good or not without quantization.

[0027] In addition, to solve the above problems, in this product inspection method, the broadband pulsed light emitted in the emission step is light having a continuous spectrum in a wavelength band of at least 1100 nm or more and 1200 nm or less.

[0028] In addition, to solve the above problems, the broadband pulsed light emitted in the emission step is light having a continuous spectrum in a wavelength band of at least 1000 nm or more and 1300 nm or less.

[0029] In addition, to solve the above problems, in this product inspection method, the slope of the change in the time of the pulsed light whose pulse has been elongated in the elongation step with respect to the wavelength is 10 picoseconds or more per 1 nm.

[0030] In addition, to solve the above problems, in this product inspection method, the light receiving step is a step of receiving the light transmitted through the product.

[0031] In addition, to solve the above problems, in this product inspection method, the irradiation step is a step of irradiating the pulsed light from one side with a pattern equal to or larger than the size of the product.

[0032] In addition, to solve the above problems, the product inspection apparatus of the present invention is a product inspection apparatus that determines whether a product is good or not by performing optical measurement, and includes:

[0033] A pulsed light source that emits broadband pulsed light;

[0034] An elongation element that elongates the pulse width of the pulsed light emitted from the pulsed light source so that the relationship between the wavelength and the elapsed time in one pulse becomes one-to-one;

[0035] A light receiver that is disposed at a position to receive the light from the product irradiated with the elongated pulsed light; and

[0036] A determination unit that processes the output data from the light receiver to determine whether the product is good or not.

[0037] In addition, to solve the above problems, in this product inspection apparatus, the determination unit is a unit that calculates the content ratio or content of a specific component of the product based on the output data from the light receiver, and compares the calculated content ratio or content with a reference value, thereby determining whether the product is good or not.

[0038] In addition, to solve the above problems, in this product inspection apparatus, the determination unit is a unit that compares, without calculating the optical characteristics, the value at a specified time within one pulse in the output data from the light receiver with a reference value, thereby determining whether it is good or not.

[0039] In addition, to solve the above problems, in this product inspection apparatus, the determination unit is a unit that compares the optical characteristics of a specified wavelength calculated based on the value at a specified time within one pulse in the output data from the light receiver or the value at the specified time with a reference value, thereby determining whether it is good or not without quantization.

[0040] In addition, to solve the above problems, in this product inspection apparatus, the pulsed light source is a light source that emits pulsed light, and the pulsed light is light having a continuous spectrum in a wavelength band of at least 1100 nm or more and 1200 nm or less.

[0041] In addition, to solve the above problems, in this product inspection apparatus, the pulsed light source is a light source that emits pulsed light, and the pulsed light is light having a continuous spectrum in a wavelength band of at least 1000 nm or more and 1300 nm or less.

[0042] In addition, to solve the above problems, in this product inspection apparatus, the stretching element is an element that performs pulsed stretching in a state where the slope of the change of time with respect to wavelength is 10 picoseconds or more per 1 nm.

[0043] In addition, to solve the above problems, in this product inspection apparatus, the light receiver is disposed at a position where it receives the light transmitted through the product.

[0044] In addition, to solve the above problems, this product inspection apparatus includes an exclusion mechanism that excludes products determined to be defective products.

[0045] In addition, to solve the above problems, in this product inspection apparatus, there is an irradiation optical system, the product is a tablet, and this irradiation optical system irradiates the tablet with the pulsed light after pulsed stretching in a pattern equal to or larger than the size of the tablet from one side.

[0046] Advantages of the Invention

[0047] As described below, the product inspection method or product inspection apparatus according to the invention of the present application irradiates light without performing a process of dissolving the manufactured product in a solution, and determines whether it is good or not based on the result. Therefore, a result of whether it is good or not can be obtained in an extremely short time, and full inspection can also be performed.

[0048] In addition, although it is a device and method for determining whether it is good or not based on the light characteristics of the target product, since pulsed light after pulse elongation is used, time-consuming operations such as scanning of a diffraction grating are not required, and high-speed determination of whether it is good or not can be performed.

[0049] Moreover, since temporal spectral splitting using the temporal wavelength uniqueness of pulsed light after pulse elongation is performed instead of spatial spectral splitting using a diffraction grating, a sufficient amount of light can be made incident on the light receiver even in a short time. Therefore, high-speed and high-SN ratio measurement can be achieved, and high-reliability determination of whether it is good or not can be performed at high speed.

[0050] In addition, when determining whether it is good or not by comparing with a reference value without calculating the light characteristics based on the value at a specified time within one pulse in the output data from the light receiver, the arithmetic processing is simplified, so that determination of whether it is good or not can be performed at a higher speed.

[0051] In addition, when the pulsed light source emits light, that is, pulsed light, having a continuous spectrum in a wavelength band of at least 1100 nm or more and 1200 nm or less, it is easier to determine whether the product is good or not.

[0052] In addition, when the pulsed light source emits light, that is, pulsed light, having a continuous spectrum in a wavelength band of at least 1000 nm or more and 1300 nm or less, it is possible to easily determine whether more types of products are good or not.

[0053] In addition, if the stretching element is an element that performs pulse stretching in such a way that the slope of the change of time with respect to wavelength becomes 10 picoseconds or more per 1 nm, sufficient wavelength resolution can be obtained, so that determination of whether it is good or not can be performed more accurately.

[0054] In addition, if the light receiver is provided at a position for receiving the light transmitted through the product, it becomes a preferable structure when determining whether the product is good or not by an absorption spectrum.

[0055] In addition, if the invention of the product inspection apparatus includes an exclusion mechanism, since the products determined to be defective are excluded, accidents in which defective products are erroneously shipped can be prevented.

[0056] In addition, in the case where the product is a tablet and the pulsed light after pulse elongation irradiates the product from one side in a pattern larger than the tablet, the SN ratio in light measurement becomes higher. Therefore, in this regard, it is possible to judge the quality of the tablet with higher accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a schematic diagram of the product inspection device of the embodiment.

[0058] Figure 2 is a schematic diagram showing pulse elongation based on an elongation element, and is a diagram schematically showing the relationship between the elapsed time and the wavelength within one pulse after pulse elongation.

[0059] Figure 3 is a schematic diagram showing an example of the dispersion characteristics of an optical fiber used as an elongation element.

[0060] Figure 4 is a schematic diagram showing the outline of the pass / fail judgment program constituting the judgment unit.

[0061] Figure 5 is a schematic diagram showing the spectrum calculation of the spectrum calculation module.

[0062] Figure 6 is a schematic diagram of the main part of the product inspection device of the second embodiment.

[0063] Figure 7 is a schematic diagram of the main part of the product inspection device of the third embodiment.

[0064] Figure 8 is a schematic diagram of the product inspection device of the fourth embodiment.

[0065] Figure 9 is a schematic diagram showing another example of the expansion element.

[0066] Figure 10 is a schematic diagram showing another example of the expansion element. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] Hereinafter, a mode (embodiment) for implementing the invention of the present application will be described.

[0068] Figure 1 is a schematic diagram of the product inspection device of the embodiment. The product inspection device of the embodiment is a device used in the production lines of various products, and is a device that judges the quality of a product by performing light measurement on the product transported to the inspection site on the production line.

[0069] The device is characterized in that it determines whether a product is good or not by optical measurement, and uses supercontinuum light as a characteristic point for the light used therefor. Further, it is characterized in that the following light is used to elongate the pulse width of the supercontinuum light which is pulsed light, and at this time, the relationship between the wavelength and the elapsed time within one pulse becomes a one-to-one relationship. The spectroscopy using a conventional diffraction grating spectrally disperses light in space, but in this embodiment, it can be said that the light is spectrally dispersed in time and utilized. Specifically, the product inspection device of the embodiment includes a pulse light source 1, an elongation element 2, a light receiver 3, and a determination unit 4.

[0070] When ultrashort pulsed light such as an ultrashort pulse laser passes through a nonlinear element such as a nonlinear optical fiber, a phenomenon of broadbanding due to nonlinear effects such as self-phase modulation occurs and is known as supercontinuum light (hereinafter, simply referred to as SC light). The pulse light source 1 of the embodiment is a light source that emits this SC light. Therefore, in this embodiment, the pulse light source 1 includes an ultrashort pulse laser 11 and a nonlinear element 12.

[0071] As the ultrashort pulse laser 11, a gain-switched laser, a microchip laser, a fiber laser, etc. can be used. In addition, as the nonlinear element 12, an optical fiber is mostly used. For example, a photonic crystal fiber or other nonlinear optical fiber can be used as the nonlinear element 12. As the mode of the optical fiber, the single-mode case is more common, but as long as the multimode also exhibits sufficient nonlinearity, it can also be used as the nonlinear element 12.

[0072] In addition, as the pulse light source 1, a light source that emits broadband light with a continuous spectrum in the near-infrared wavelength band of about 1100 to 1200 nm is preferably used. In this embodiment, as described later, the quality of the product is determined by the absorption spectrum in the near-infrared region. From the viewpoint of increasing the range of products that can be inspected, more preferably, the pulse light source 1 emits broadband light with a continuous spectrum in the wavelength band of at least about 1000 to 1300 nm.

[0073] The elongation element 2 is an element that elongates the pulse width of the SC light emitted from the pulse light source 1. The light emitted from the pulse light source 1 has a broadened wavelength band, but as the pulse width, it maintains a short pulse in the femtosecond to nanosecond range. In this state, it is difficult to use for optical measurement, so the pulse is elongated by the elongation element 2. At this time, the important point is to elongate it in such a way that the relationship between the wavelength and the elapsed time within one pulse becomes one-to-one.

[0074] Specifically, in the present embodiment, as the elongation element 2, a wavelength dispersion optical fiber such as that used as a dispersion compensating fiber (DCF) in the optical communication field is used. The wavelength dispersion optical fiber is an optical fiber having sufficient group delay characteristics and is classified into a normal dispersion optical fiber and an anomalous dispersion optical fiber. Either can be used, but in this embodiment, a normal dispersion optical fiber is used as the elongation element 2.

[0075] Figure 2 FIG. is a schematic diagram showing pulse elongation based on the elongation element 2, and is a diagram schematically showing the relationship between the elapsed time and the wavelength within one pulse after pulse elongation. Figure 2 (1) represents the intensity with respect to the elapsed time within one pulse, Figure 2 (2) represents the intensity of each wavelength. Additionally, Figure 2 (3) represents the relationship between the elapsed time and the wavelength within the pulse.

[0076] As Figure 2 (1) - (3) show, the elapsed time and the wavelength within one pulse of the SC light after pulse elongation correspond one-to-one. That is, if the start time of one pulse is set as t 1 , and the end time of this pulse is set as t n , then at the initial stage of the start of one pulse, there is light with the longest wavelength λ 1 . As time passes, the wavelength of the existing light shifts towards the short wavelength side. And, there is light with the shortest wavelength λ n near the end stage t n of the pulse. Thus, the wavelength and the elapsed time within the pulse correspond one-to-one. Therefore, if the elapsed time from the initial stage t 1 of the pulse is determined and the intensity of the light is obtained, then the intensity becomes the intensity of a specific wavelength. That is, the intensity at each elapsed time is the intensity of each wavelength, and it is not limited to the spectral spectrum.

[0077] Such uniqueness of the elapsed time and the wavelength is particularly important in the structure of the embodiment where the quality of the product is judged by obtaining the spectral spectrum from the output change with time from the light receiver 3, but the ratio of the wavelength with respect to the passage of time is also important. This is the degree of time deviation (time dispersion) for a difference of 1 nm in wavelength, which is represented by Δt / Δλ in Figure 2 (3). In the embodiment, when Δλ is 1 nm, Δt is 10 picoseconds or more.

[0078] Each characteristic of such an elongation element 23 can be achieved by selecting an optical fiber having appropriate dispersion characteristics. Hereinafter, this will be described.

[0079] Figure 3It is a schematic diagram showing an example of the dispersion characteristics of the optical fiber used as the elongation element 2. The optical fiber as the elongation element 2 preferably does not include zero dispersion at least in the wavelength band of 1100 to 1200 nm. That is, in the wavelength band of 1100 to 1200 nm, it is preferably all normal dispersion characteristics or all anomalous dispersion characteristics. Figure 3 It is an example of an optical fiber with normal dispersion characteristics.

[0080] If zero dispersion is included in the wavelength band of 1100 to 1200 nm, it corresponds to two or more wavelengths at the same time, and an unwanted nonlinear optical effect is likely to occur. Therefore, the above-mentioned time-wavelength uniqueness may be destroyed. In addition, compared with anomalous dispersion, an optical fiber showing normal dispersion in 1100 to 1200 nm is preferred. SC light mostly has the characteristic that the light on the long wavelength side exits first and the light on the short wavelength side exits later. In this case, the pulse of SC light has light of long wavelength at the beginning, and then shifts to the short wavelength side as time passes. If an optical fiber with normal dispersion characteristics is used as the elongation element 2, as a result, the light on the short wavelength side is further delayed compared with the light on the long wavelength side. Therefore, the above-mentioned time relationship is maintained and it becomes a further elongated state. Therefore, long-pulse light can be easily obtained without destroying the time-wavelength uniqueness.

[0081] In addition, an optical fiber with anomalous dispersion can also be used as the elongation element 2 in the wavelength band of 1100 to 1200 nm. In this case, since the light on the long wavelength side existing at the beginning of the pulse in SC light is delayed and the light on the short wavelength side existing at a later time is dispersed while advancing, the time relationship within one pulse is reversed. At the beginning of one pulse, there is light on the short wavelength side, and pulse elongation occurs in a state where there is light on a longer wavelength side as time passes. However, compared with the case of normal dispersion, the propagation distance for pulse elongation needs to be further extended in many cases, and the loss is likely to become larger. Therefore, normal dispersion is preferred in this regard.

[0082] As Figure 1 shown, in order to irradiate the product P to be judged as good or bad with the SC light (hereinafter referred to as elongated SC light) after pulse elongation, the product inspection device of the embodiment includes an irradiation optical system 5. In addition, in order to arrange the product P at the irradiation position of the elongated SC light based on the irradiation optical system, a arranging member 6 is provided.

[0083] The irradiation optical system 5 is configured to irradiate the elongated SC light from above, and a receiving plate 61 is used as the arranging tool 6. In this embodiment, since the good or bad is judged by transmitted light, the receiving plate 61 is transparent within the measurement wavelength range.

[0084] In addition, the irradiation optical system 5 includes a beam expander 51. The elongated pulse SC light is a laser, and is light with a small beam pattern. This is because the elongated pulse light is irradiated from one side of the product P, but for performing a higher-precision inspection, it should be irradiated from one side in a pattern equal to or larger than the size of the product P.

[0085] The light receiver 3 is disposed at a position for receiving the transmitted light from the product P irradiated with the elongated SC light. The light receiver 3 uses a light receiver having sufficient sensitivity within the measurement wavelength range. For example, a highly sensitive element having a photoelectric conversion element such as an InGaAs photodiode is used as the light receiver 3.

[0086] The determination unit 4 is a unit that processes the output data from the light receiver 3 to determine whether the product is good or not. As the determination unit 4, a general-purpose PC is used in the present embodiment. The general-purpose PC includes a processor 41 and a storage unit (hard disk, memory, etc.) 42. A pass / fail determination program 43 for processing the output data from the light receiver 3 and outputting a pass / fail determination result and other necessary programs are installed in the storage unit 42. In addition, an AD converter 7 is provided between the light receiver 3 and the determination unit 4, and the output data from the light receiver 3 is converted into digital data and input to the determination unit 4.

[0087] Figure 4 FIG. is a diagram showing an outline of the pass / fail determination program 43 constituting the determination unit 4. The ultimate purpose of the pass / fail determination program 43 is to determine whether the product is good or not based on the data output from the light receiver 3 (hereinafter, simply referred to as output data). Therefore, this method is roughly divided into two types. One is a method of calculating a spectrum based on the output data and comparing it with a reference value to determine pass / fail. The other is a method of directly comparing with the reference value without calculating the spectrum from the output data to determine pass / fail. In the present embodiment, the former method is used.

[0088] As Figure 4 shown, the pass / fail determination program 43 includes a spectrum calculation module 431, a spectrum quantization module 432, and a determination module 433. The spectrum calculation module 431 is a module that processes the output data to calculate the absorption spectrum S 2 . The spectrum quantization module 432 is a module that obtains a quantity (hereinafter, referred to as a quantization value) Q that can be compared with a reference value based on the calculated absorption spectrum S 2 . The determination module 433 is a module that compares the calculated quantization value Q with the reference value, determines pass / fail, and outputs the result as the execution result of the program.

[0089] First, the spectrum calculation module 431 will be described. Figure 5 FIG. is a schematic diagram showing the spectrum calculation based on the spectrum calculation module 431.

[0090] As described above, in the stretched SC light, the wavelength within the pulse and the elapsed time correspond one-to-one. Therefore, the spectral calculation module 431 first converts the horizontal axis of the output data D from time to wavelength. The output data D is a certain data set, which is the value v at each moment t of each specified period Δt 1 、t 2 、t 3 、... The spectral calculation module 431 recaptures the value at each moment as the corresponding wavelength λ 1 、v 2 、v 3 、... The spectral calculation module 431 recaptures the value at each moment as the corresponding wavelength λ 1 、λ 2 、λ 3 、... as the measured spectrum S1.

[0091] Next, the reference spectral data S 0 is applied to calculate the absorption spectrum S 2 . That is, the reference spectral data S 0 is obtained in advance by irradiating the stretched pulse light in a state where nothing is disposed on the receiving plate 61 and making it incident on the light receiver 3, and is provided to the spectral calculation module 431 as a constant.

[0092] The reference spectral data S 0 is also a certain data set, which is a set of intensities (reference intensities) V at each moment t 1 、t 2 、t 3 、... The spectral calculation module 431 calculates v 1 、V 2 、V 3 、... for each wavelength λ 1 、λ 2 、λ 3 、... calculates v 1 / V 1 、v 2 / V 2 、v 3 / V 3 、... and takes the logarithm of the reciprocal as needed as the absorption spectrum S 2 .

[0093] In addition, although not shown in the figure, when the light receiver 3 has a sensitivity characteristic with respect to the wavelength (when the photoelectric conversion characteristic between wavelengths is not smooth), correction corresponding to the sensitivity characteristic is performed. Coefficients for correction are preset, and the values at each moment t 1 、t 2 、t 3 、... are multiplied by the coefficients, and used as the values at each wavelength λ 1 、λ2 and the intensity v in λ 3 ,..., the measurement spectrum S is obtained 1 v 2 v 3 v,..., 1 .

[0094] The calculated absorption spectrum S 2 is the sum of the absorption spectra of the respective components contained in the product. Although it is also possible to judge whether the product is good or not based on the amounts of all these contained components, since it is too complicated, it is judged whether it is good or not based on the amount of a certain specific component. The certain specific component is the component that has the greatest influence on the quality of the product, or the most abundant component in the product. In the case of pharmaceuticals, there are also cases where it is judged whether it is good or not based on the amount of the active ingredient.

[0095] In short, in the present embodiment, the absorption spectrum S in the near-infrared region is used 2 to judge whether it is good or not. As is well known, in the near-infrared region, the absorption bands of many materials overlap, and it is difficult to directly obtain the amount of the target component from the calculation results of the absorption spectrum. Therefore, the spectral quantification module 432 adopts a chemometric method.

[0096] Regarding chemometrics, methods such as PCA (principal component analysis), PCR (principal component regression analysis), and PLSR (partial least square regression, PLS regression) analysis are known. Although any method can be adopted, as an example, the case of performing PLSR will be described.

[0097] In the case of performing PLSR, a plurality of samples (products) with known amounts of the target component are measured in the same way to obtain a data set. Then, regression analysis is performed based on the obtained multiple data sets to obtain regression coefficients. In actual quantification, the obtained regression coefficients are used to predict the amount of the target component, and the predicted value is used as the quantification value.

[0098] PLSR is a method developed from PCA or PCR, and principal component analysis is first performed. That is, as shown in Equation 1 below, the multivariate data X (here, the absorption spectrum measured for a sample with a known amount of the target component) is decomposed into principal component scores T, loading vectors R, and residuals E.

[0099] [Quantity 1]

[0100] X = TR + E (Equation 1)

[0101] In PLSR, principal component analysis is performed on the multi-variable data X. To avoid collinearity, the value of the principal component score T obtained here is used for regression analysis. At this time, only the part related to the amount of the principal component in the spectral data set X is taken out, and the regression coefficient is obtained by the least squares method. Then, a calibration curve is made according to the regression coefficient thus obtained. Regarding chemometrics other than PLSR, it is described in Non-Patent Document 1 or other documents, so further description is omitted.

[0102] As Figure 4 shown, after the spectral quantification module 432 is executed, the pass / fail judgment program 43 executes the judgment module 433. The judgment module 433 is a module that judges pass / fail by comparing the quantization value Q obtained by the spectral quantification module 432 with a reference value. For the judgment module 433, the reference value and the tolerance of the deviation from the reference value are given in the form of constants. The judgment module 433 judges pass / fail based on these and outputs the result as the execution result of the pass / fail judgment program 43.

[0103] It should be noted that the amount Q of the target component sometimes exists in the form of a ratio (including ratio) to the whole, and sometimes exists in the form of an absolute value (content). In the case of calculating the absolute value, a calibration curve is made in a way that can calculate the absolute value, or in the case of a weight ratio, the weight of the product is measured and calculated separately.

[0104] And, actually, preprocessing such as smoothing or second-order differentiation is performed on the output data D, and then the regression coefficient obtained by PLSR is applied to obtain the quantization value Q. At this time, in order to extract only the part related to the target component, wavenumber domain selection is performed, and the quantization value is obtained thereon.

[0105] Next, use Figure 1 and Figure 4 to explain the overall operation of the product inspection device of the above embodiment. The following explanation is also an explanation of an embodiment of the invention of the product inspection method.

[0106] As described above, the product inspection device is arranged at the inspection site of the product production line. The product P is transported to the inspection site. Sometimes it is transported by a handling mechanism such as a conveyor, and sometimes it is carried by an operator by hand.

[0107] The product P is arranged on the receiving plate 61. This may also be automatically arranged by a mechanism such as a robot, or arranged by an operator.

[0108] The SC light emitted from the pulsed light source 1 is pulse-stretched by the stretching element 2 with time-wavelength uniqueness to become stretched SC light. The stretched SC light is irradiated onto the product P through the irradiation optical system 5. The stretched SC light transmitted through the product P reaches the light receiver 3 and is photoelectrically converted.

[0109] The output from the light receiver 3 becomes a digital signal through the AD converter 7 and is input as output data D to the determination unit 4. In the determination unit 4, a pass / fail determination program 43 is executed. The pass / fail determination program 43 calculates the absorption spectrum S2 based on the output data D, quantifies it through PLSR, and compares the value Q with a reference value to determine pass / fail. The pass / fail determination result is stored in the storage unit 42 within the determination unit 4.

[0110] According to the product inspection apparatus and product inspection method of such an embodiment, for the manufactured product, light is directly irradiated without performing a process such as dissolving in a solution, and pass / fail is determined based on the result. Therefore, good results can be obtained in an extremely short time, and full inspection can also be performed.

[0111] In addition, although it is an apparatus and method for determining pass / fail based on the spectroscopic characteristics of the product, since stretched SC light is used, there is no need for a time-consuming operation such as changing the posture of the diffraction grating for wavelength scanning, and high-speed and good determination can be performed.

[0112] Moreover, since it is not spatial spectroscopy using a diffraction grating but temporal spectroscopy utilizing the time-wavelength uniqueness of stretched SC light, a sufficient amount of light can be made to enter the light receiver 3 even in a short time. In the case of spatial spectroscopy using a diffraction grating, since there is a loss during spatial dispersion, the light incident on the light receiver easily becomes weak, and in order to perform high-SN ratio measurement, it is necessary to make the light incident for a long time. In the apparatus and method of the embodiment, a sufficient amount of light can be made to enter the light receiver 3 even in a short time, so high-speed and high-SN ratio measurement can be performed, and high-reliability pass / fail determination can be performed at high speed.

[0113] In addition, in the above description, it is described that pass / fail determination is performed using the stretched SC light of one pulse. However, in order to ensure the necessary light amount, sometimes the stretched pulsed light of multiple pulses is made to enter the light receiver 3, and pass / fail determination is performed based on the sum or average of the values at each moment.

[0114] In the above-described embodiment, the fact that the pulsed light source 1 emits light having a continuous spectrum within a wavelength band of 1100 nm or more and 1200 nm or less, that is, supercontinuum light, has the significance of making it easy to determine whether the product is good or not. In most cases, the near-infrared region of 1100 to 1200 nm has an absorption spectrum corresponding to the components contained in the product, and analytical techniques such as chemometrics are also well-developed. Therefore, it becomes easy to determine good or bad.

[0115] In addition, even in the near-infrared region, the absorption spectra are often somewhat different for different products. Considering this, it is more preferable that the pulsed light source emits SC light having a spectrum within a wavelength band of 1000 nm or more and 1300 nm or less. This is because it is possible to easily determine whether various products having absorption spectra in the near-infrared region are good or not.

[0116] In addition, in terms of the pulse stretching of the stretching element 2 such that the slope of the change of time with respect to wavelength becomes 10 picoseconds or more per 1 nm, it has the significance of improving the quality of the determination of good or bad from the viewpoint of wavelength resolution. As can be seen from the above description, if the slope of the change of time with respect to wavelength becomes small, the wavelength resolution (resolution when dividing wavelengths in time) becomes low due to the relationship with the responsiveness of the light receiver. If this is the case, the determination of good or bad is made based on values at wavelengths with a rather large jump during that period, and the accuracy of the determination of good or bad is likely to decrease. If the slope of the change of time with respect to wavelength is 10 picoseconds or more per 1 nm, there is no such problem.

[0117] In addition, the structure in which the light receiver 3 is provided at a position for receiving the light transmitted through the product is a more preferable structure in terms of determining whether the product is good or bad by absorption spectrum. In the case of determining good or bad using the absorption spectrum, it is also possible to capture the reflected light from the product irradiated with light to calculate the absorption spectrum, but in the case of large surface reflection, the measurement accuracy is likely to become low, and it may be difficult to determine good or bad. In addition, mainly the absorption spectrum caused by the product surface is obtained, so it is difficult to obtain information inside the product. In the case where the light receiver 3 is provided at a position for receiving the light transmitted through the product, there is no such problem.

[0118] In addition, irradiating the product with stretched SC light in a pattern of the same size or larger than the product also has the significance of improving the SN ratio and the accuracy of the determination of good or bad. As the output data required for the determination of good or bad, even the data obtained by photoelectrically converting the light emitted from a part of the area of the product estimated from the light receiver 3 reflects the properties of the product and can be determined well. However, since background light and the like also enter the light receiver 3 and become noise, the light (signal light) from the product should be increased as much as possible to improve the SN ratio, and the structure of the embodiment satisfies this requirement.

[0119] Next, the product inspection apparatus and product inspection method of the second embodiment will be described.

[0120] Figure 6 It is a schematic diagram of the main part of the product inspection apparatus of the second embodiment. In the second embodiment, the pass / fail judgment program 43 included in the judgment unit 4 is different from that of the first embodiment. The second embodiment is a device and method for performing pass / fail judgment based on the output data from the light receiver 3, but without performing the calculation of light characteristics as in the first embodiment. That is, the pass / fail judgment program 43 in the second embodiment includes a direct evaluation module 434 that obtains a value by evaluating the output data from the light receiver 3 without calculating light characteristics by chemometrics, and the judgment module 433 judges pass / fail based on the value obtained in the direct evaluation module 434.

[0121] Specifically, in the second embodiment, the output data D is also input from the light receiver 3 to the judgment unit 4 via the AD converter 7. This is also composed of the values v at times t 1 、t 2 、t 3 、... 1 、v 2 、v 3 、...

[0122] In the second embodiment, for this data set D, a calibration curve is also prepared in advance by chemometrics. However, at this time, the calibration curve is prepared without calculating light characteristics such as absorption spectra. That is, multiple products with known amounts of the target component are irradiated with stretched pulse light in the same manner, and multiple data sets D (v 1 、v 2 、v 3 、...) are obtained from the light receiver 3. Then, regression processing such as PLSR is directly performed using the data set D (v 1 ,v 2 ,v 3 ,...) to calculate regression coefficients and obtain the calibration curve.

[0123] Therefore, as Figure 6 shown, in the second embodiment, when stretched SC light is irradiated on a certain target product to obtain the output data D (v 1 、v 2 、v 3 、...), the regression coefficients are applied to calculate the amount Q' of the target component, and it is judged whether it is within the allowable range with respect to the reference value. Moreover, if it is within the allowable range, it is a qualified product, and if it is outside the allowable range, it is a defective product.

[0124] In the second embodiment, since there are no steps or program modules for calculating optical characteristics, the pass / fail judgment program 43 is simplified, and the pass / fail judgment can be performed at a higher speed. Additionally, even when the light receiver 3 has a sensitivity characteristic with respect to wavelength, as long as the same light receiver 3 is used, correction corresponding to the sensitivity characteristic is not required. The structure is also simplified in this regard. However, in the case where the sensitivity of the light receiver 3 changes over time, periodic correction is required.

[0125] In addition, in the case of the first embodiment, the result of the stage where the spectral calculation module 431 is executed can be output additionally. This result is the absorption spectrum of the target product, but in the case where the absorption spectrum needs to be known for product evaluation or other purposes, the first embodiment is preferred.

[0126] In addition, in the above first embodiment, the pass / fail judgment program 43 calculates the absorption spectrum S2 from the output data D and quantifies it by PLSR, but there may be cases where the pass / fail judgment is performed without quantification. That is, in the case of a reference absorption spectrum which is the absorption spectrum determined in advance as a qualified product, after the spectral calculation module 431 calculates the absorption spectrum S2, there may be cases where the judgment module 433 performs the pass / fail judgment by comparing this absorption spectrum S2 with the reference absorption spectrum.

[0127] In addition, as another embodiment, there may be cases where the pass / fail is judged only by the optical characteristics or the photoelectric conversion value for a certain specific wavelength. For example, a certain product contains a certain characteristic component, and the component has strong absorption at a certain specific wavelength. Moreover, other components do not absorb at this wavelength. In this case, there may also be cases where: in the output data, the data corresponding to this wavelength at a certain moment is obtained, and the pass / fail judgment is performed only by comparing this data or the optical characteristics calculated from this data with a reference value.

[0128] Next, the product inspection device of the third embodiment will be described. Figure 7 It is a schematic diagram of the main part of the product inspection device of the third embodiment.

[0129] The third product inspection device includes an exclusion mechanism 8 that excludes the products judged as defective by the judgment unit 4 from the production line. Regarding the exclusion mechanism 8, it can be appropriately configured according to the shape and size of the product, and an example is shown in Figure 7 One example is shown. Figure 7 (1) is a front schematic view, and (2) is a top schematic view.

[0130] In this embodiment, as the product P, a small flat product such as a tablet is envisioned. In the first and second embodiments, the receiving plate 61 is used as a component, but in this embodiment, a rotating drum 62 is provided as a component.

[0131] The rotating drum 62 is a mechanism that is concentric with the rotating drum 62 and rotates around a horizontal rotation axis 621. Although not shown in the figure, it is configured to be connected to a belt and a motor with respect to the horizontal central axis and rotate at a prescribed rotational speed.

[0132] The rotating drum 62 has a structure that holds a plurality of products P on its circumferential surface. On the circumferential surface of the rotating drum 62, suction holes (not shown) are provided at equal intervals. Each suction hole communicates with a vacuum pump (not shown), and the discharging mechanism 8 includes a vacuum suction system 81 that independently opens and closes the suction of each suction hole. As Figure 7 shown, each product P is conveyed by a conveyor or the like and abuts against the circumferential surface of the rotating drum 62, and is adsorbed and held by the suction holes. As the vacuum suction system 81, a mechanism that independently opens and closes the communication between each suction hole and the vacuum pump can be adopted. For example, a mechanism in which an opening and closing plate is provided for each suction hole and each opening and closing plate is independently driven can be adopted.

[0133] In addition, each product P is adsorbed in such a manner that its thickness direction faces horizontally. As a structure for this, for example, a groove is provided on the conveyor, and the width of the groove is set to the thickness of the product. Each product falls into the groove and is conveyed, and at this time, it assumes a vertically standing posture (the radial direction assumes a posture within a vertical plane), and is adsorbed and held by the rotating drum 62 in this posture.

[0134] On the other hand, with respect to the rotating drum 62, a light irradiation unit and a light receiving unit are provided in a prescribed positional relationship. The light irradiation unit is a part of the irradiation optical system 5, but in this embodiment, it is the irradiation side optical fiber 52 that guides and emits the light from the light source 1. The light receiving unit is the light receiving side optical fiber 31 into which the transmitted light from the product P is incident. A light receiver is arranged at a position where the light emitted from the output side of the light receiving side optical fiber 31 is received.

[0135] The discharging mechanism 8 includes a control unit (not shown) that controls the vacuum suction system 81, and the output of the determination unit 4 is input to the control unit. In addition, for the rotating drum 62, the position where the product is released in the case of a non-defective product and the position where it is released in the case of a defective product are set to different positions. A conveyor or the like for transporting to the next process is arranged at the position where the non-defective product is released, and a disposal inlet is arranged at the position where the defective product is released.

[0136] When the control unit outputs a determination that the product is defective from the determination unit 4, it disconnects the vacuum suction of the suction hole that has adsorbed the product at the position where the defective product is released, and in other cases, disconnects it at the position where the non-defective product is released.

[0137] According to the third embodiment, since the rejection mechanism 8 is provided, products determined to be defective can be reliably and automatically rejected from the production line. Therefore, accidents in which defective products are mistakenly shipped are prevented.

[0138] As the structure of the rejection mechanism 8, various structures can be adopted in addition to the above. For example, a conveyor for conveying products has an opening / closing part such as an opener / closer, and is configured to convey products by placing them on each opening / closing part. For products determined to be defective, a structure that opens the opening / closing part and causes them to fall, or a structure that rejects them from the production line by picking them up with a robotic arm or discharging them with air blow can be adopted.

[0139] In addition, in each of the above embodiments, there may be a case where the quality of products that move for conveyance or the like is judged without stopping their movement. It is also possible to irradiate a certain area with elongated SC light in advance, and cause the products to move through this area, and judge the quality at this time. If multiple pulses are received during the process of passing through the irradiation area, averaging is performed in the processing of the output data of the light receiver 3 to judge the quality. It is possible to judge the quality without stopping the movement, and to judge the quality without reducing the productivity, and its significance is particularly remarkable in the case of performing 100% inspection.

[0140] In addition, although the present invention can perform 100% inspection, it is not necessarily required to perform 100% inspection. Even in the case of performing only sampling inspection, the configuration of the present invention that can perform high-reliability inspection at high speed contributes greatly to manufacturing high-quality products with high productivity.

[0141] Next, the product inspection apparatus of the fourth embodiment will be described. Figure 8 It is a schematic diagram of the product inspection apparatus of the fourth embodiment.

[0142] As Figure 8 shown, in the spectroscopic measurement apparatus of the fourth embodiment, a branching element 52 that branches the pulsed light elongated by the pulse elongation element 2 is provided. As the branching element 52, a beam splitter is used in the present embodiment.

[0143] The branching element 52 divides the optical path from the pulse light source 1 into a measurement optical path and a reference optical path. On the measurement optical path, a receiving plate 3 is arranged in the same manner as in the first embodiment, and a measurement light receiver 3 is arranged at a position where the light transmitted through the object P on the receiving plate 61 is received.

[0144] A reference light receiver 31 is arranged on the reference optical path. On the reference light receiver 31, the light branched by the branching element 52 and advancing on the reference optical path is directly incident. This light (reference light) is the light that is incident on the reference light receiver 31 without passing through the object P and the reference spectral data is obtained in real time.

[0145] The measurement light receiver 3 and the reference light receiver 31 are respectively connected to the arithmetic unit 4 via AD converters 7 and 71. The pass / fail judgment program 43 in the arithmetic unit 4 is programmed to perform real-time reference of the reference intensity spectrum. That is, the measurement values v 1 、t 2 、t 3 、… at each time t are input from the measurement light receiver 3, and the reference intensities V 1 、V 2 、V 3 、... at the same times t are input from the reference light receiver 31. The pass / fail judgment program 43 calculates v 1 、t 2 、t 3 、... at each time t and the reference intensities V 1 、V 2 、V 3 、... (reference spectral data) at the same times t. The pass / fail judgment program 43 calculates v 1 、t 2 、t 3 、... and the wavelengths λ 1 、λ 2 、λ 3 、... according to the relationship between the times t 1 / V 1 、v 2 / V 2 、v 3 / V 3 、... obtained in advance during one pulse, and takes the logarithm of the reciprocal as needed to obtain the absorption spectrum. In the case of measuring the reflection spectrum and the scattering spectrum, it can also be performed in the same way with the reference spectral data obtained in real time.

[0146] In this embodiment, since the reference spectral data is obtained in real time, the acquisition of the reference spectral data at regular intervals is not performed. Except for this point, it is the same as the first embodiment.

[0147] According to the fourth embodiment, it is not necessary to separately obtain reference spectral data, so the efficiency of the overall pass / fail judgment operation is improved. Additionally, in the first embodiment, when the characteristics of the pulsed light source 1 and the stretching element 2 are likely to change, calibration operations need to be performed frequently, but this is not required in the fourth embodiment. Even if the characteristics of the pulsed light source 1 and the stretching element 2 do not change, calibration operations may be required when the measurement environment is different (for example, when temperature conditions, backlight conditions, etc. are different). In the fourth embodiment, calibration operations are not required in such cases either, so the inspection efficiency is relatively high. However, in the fourth embodiment, the light beam from the pulsed light source 1 is split into two, so the light beam that can irradiate the object P can be correspondingly reduced. Therefore, when it is necessary to irradiate the object P with a higher intensity for inspection, the first embodiment is advantageous.

[0148] Next, with reference to Figure 9 and Figure 10 other examples of the stretching element 2 will be described. Figure 9 and Figure 10 are schematic diagrams showing other examples of the stretching element 2. As the stretching element 2, in addition to optical fibers, diffraction gratings, chirped fiber Bragg gratings (CFBGs), prisms, etc. can also be used. For example, as shown in Figure 9 (1), two diffraction gratings 21 can be used to disperse the wavelength. When the light is reflected back by the mirror, wavelength dispersion is performed by the two diffraction gratings 21 in the forward and return paths (a total of 4 times). Thereby, an optical path difference is imparted according to the wavelength, and pulse stretching is performed in a state where time-wavelength uniqueness is achieved. In this example, the optical path becomes shorter for longer-wavelength light.

[0149] Additionally, as shown in Figure 9 (2), a CFBG 22 can also be used for pulse stretching. An FBG is an optical fiber in which portions with refractive index changes are periodically provided in the longitudinal direction of the core to form a diffraction grating. Among them, the CFB 22 can cause the reflection position to be different according to the wavelength in such a way that the function of a chirped mirror can be realized using an optical fiber. When used as the pulse stretching element 2, in the CFBG 22, among the incident light, for example, light on the long-wavelength side is reflected and returned near the front side of the traveling direction in the optical fiber, and an refractive index variation layer in the core is formed in such a way that it is reflected and returned on the inner side as it becomes the short-wavelength side. In this case, similar to a normal dispersion optical fiber, the shorter the wavelength side, the more delayed the return, so time-wavelength uniqueness can be ensured.

[0150] And, as shown in Figure 9As shown in (3), a prism 23 can also be used for pulse stretching. In this example, four prisms (two pairs of prisms) 23 are used, and the optical path is shorter on the longer wavelength side and longer on the shorter wavelength side, thereby constituting the stretching element 2. In this example, the light reaches the light receiver 3 with a greater delay on the shorter wavelength side, so that the time-wavelength uniqueness can be ensured.

[0151] In addition, in Figure 9 the examples of (1) to (3), an optical path difference is formed when the light is reflected back. As the structure for extracting the light of the loop, a structure in which a structure obtained by combining a polarization beam splitter and a quarter-wave plate is arranged on the optical path in front of the stretching element 2 can be adopted. Regarding the forward path, the light travels in the order of the polarization beam splitter and the quarter-wave plate and is incident on the pulse stretching element 2. Regarding the loop, the light returning from the pulse stretching element 2 travels in the order of the quarter-wave plate and the polarization beam splitter.

[0152] In addition, in Figure 10 an example of using a plurality of optical fibers 25 as the stretching element 2 is shown. In this example, the pulsed light is split into lights of respective wavelengths by an arrayed waveguide grating (AWG) 24, and the lights of respective wavelengths are pulse-stretched by the respective optical fibers 25.

[0153] The arrayed waveguide grating 24 is constituted by forming respective functional waveguides 242 to 246 on a substrate 241. The respective functional waveguides are a plurality of arrayed waveguides 242 having slightly different optical path lengths, slab waveguides 243 and 244 connected to both ends (the incident side and the output side) of the arrayed waveguide 242, an incident-side waveguide 245 for making light incident on the incident-side slab waveguide 243, and respective output-side waveguides 246 for extracting lights of respective wavelengths from the output-side slab waveguide 244.

[0154] The slab waveguides 243 and 244 are free spaces, and the light incident through the incident-side waveguide 245 expands in the incident-side slab waveguide 243 and is incident on the respective arrayed waveguides 242. Since the lengths of the respective arrayed waveguides 242 are slightly different, the phases of the lights reaching the terminals of the respective arrayed waveguides 242 are displaced (shifted) by this difference respectively. The light diffracts and exits from the respective arrayed waveguides 242, but the diffracted lights pass through the output-side slab waveguide 244 while interfering with each other and reach the incident ends of the output-side waveguides 246. At this time, due to the phase shift, the intensity of the interfering light is the highest at the position corresponding to the wavelength. That is, lights having different wavelengths in sequence are incident on the respective output-end waveguides 246, and the lights are spatially split. Strictly speaking, the respective output-side waveguides 246 are formed such that the respective incident ends are located at the positions thus split. Each optical fiber 25 serving as the stretching element 2 is connected to each output-side waveguide 246.

[0155] Each optical fiber 25 may use the same optical fiber or optical fibers with different characteristics. Since light with different wavelengths is incident on each optical fiber 25 in sequence, it is preferable to make the lengths of the optical fibers 25 different according to the wavelengths. By adjusting the lengths of the optical fibers 25, the delay of each wavelength is adjusted to obtain appropriate time-wavelength uniqueness.

[0156] In addition, the multiple optical fibers 25 may also be a bundle of optical fibers. Additionally, multi-core optical fibers may be used to perform pulse stretching in such a way that light of each wavelength is transmitted in each core.

[0157] In the above description, tablets are exemplified as products, but as long as it is a product to which a quality-imparting component is added, the present invention can be applied to any product. As oral products having shapes such as tablets and granules, in addition to pharmaceuticals, various foods such as health foods and supplements can be cited, and the present invention can also be applied to various industrial products for which the amount of the added component after completion may be a problem. For example, for fine components or electronic components manufactured by semiconductor processes, it is possible to consider making a judgment of good or bad according to the present invention during or after manufacturing.

[0158] The above examples are examples of solid-phase products, but the present invention can also be applied to liquid-phase products. For example, it can be used for the inspection of liquid-phase pharmaceuticals such as oral liquids. Specifically, an example can be cited in which pulse light is irradiated on the manufactured liquid-phase pharmaceutical through a transparent container to check good or bad. Additionally, there are cases where various reagents for research or synthesis are products, and the present invention can also be applied to liquid-phase reagents.

[0159] In addition, as the pulse light source 1, in addition to a light source that emits SC light, an ASE (Amplified Spontaneous Emission) light source, an SLD (Super luminescent diode) light source, etc. can also be adopted. Since the ASE light source is light generated in an optical fiber, when an optical fiber is used as the pulse stretcher 2, the affinity is high, and broadband pulse light can be incident on the stretching element 2 with low loss, and broadband pulse light can be efficiently stretched. In addition, since the SLD light source can also extract light emission from a narrow active layer, it can be incident on the extension element 2 with low loss, and broadband pulse light can be efficiently stretched.

[0160] Symbol Explanation

[0161] 1 Pulse light source

[0162] 11 Ultra-short pulse laser

[0163] 12 Nonlinear element

[0164] 2 Stretching element

[0165] 3 Light receivers

[0166] 4 Judgment unit

[0167] 43 Good or bad judgment program

[0168] 431 Spectrum calculation module

[0169] 432 Spectrum quantization module

[0170] 433 Judgment module

[0171] 5 Irradiation optical system

[0172] 51 Beam expander

[0173] 6 Fitting

[0174] 61 Receiving plate

[0175] 8 Exclusion mechanism.

Claims

1. A product inspection method, which determines whether a product is good or not by performing optical measurement, characterized in that: have: An emission step, wherein continuous pulsed light in a wide-band spectrum is emitted from the light source; an elongation step of elongating the pulse width of the emitted pulse light by means of an elongation element so that the relationship between the wavelength and the elapsed time in one pulse becomes one to one; An irradiation step of irradiating the pulse light stretched in the stretching step onto the irradiation area; Moving process, where the product moves through the irradiation area without stopping; A light receiving step in which a light receiver receives transmitted light from a product irradiated with the elongated pulse light; as well as The judgment process processes the output data from the light receiver to judge whether the product is good or not. The stretching process uses an arrayed waveguide diffraction grating to split the pulse light into lights of each wavelength, and uses each optical fiber to stretch the pulses of the lights of each wavelength. The optical fiber has different lengths depending on the wavelength. The judgment process is a process of calculating the content ratio or content of a specific component of the product based on the output data, and comparing the calculated content ratio or content with a reference value to thereby judge whether the product is good or not.

2. The product inspection method according to claim 1, characterized in that: The determination step is a step of determining whether the output data from the optical receiver is good or not by comparing the value at a predetermined time point in one pulse with a reference value without calculating the light characteristics.

3. The product inspection method according to claim 1, characterized in that: The determination step is a step of determining whether the output data from the optical receiver is good or bad by comparing the optical characteristic of a predetermined wavelength calculated based on the value at a predetermined time in one pulse of the output data from the optical receiver or the value at a predetermined time with a reference value.

4. The product inspection method according to any one of claims 1 to 3, characterized in that: The broadband pulse light emitted in the emission step is light having a continuous spectrum in a wavelength band of at least 1100 nm to 1200 nm.

5. The product inspection method according to any one of claims 1 to 3, characterized in that: The broadband pulse light emitted in the emission step is light having a continuous spectrum in a wavelength band of at least 1000 nm to 1300 nm.

6. The product inspection method according to any one of claims 1 to 3, characterized in that: The slope of the change in time with respect to wavelength of the pulse light subjected to pulse extension in the extension step is 10 picoseconds or more per nm.

7. The product inspection method according to any one of claims 1 to 3, characterized in that: The irradiation step is a step of irradiating the pulse light from one side in a pattern having a size equal to or larger than that of the product.

8. A product inspection device that determines whether a product is good or not by performing optical measurement, characterized in that: have: Pulse light source, emitting broadband pulse light; an elongation element for elongating the pulse width of the pulse light emitted from the pulse light source in such a manner that the relationship between the wavelength and the elapsed time in one pulse becomes one to one; an irradiation optical system for irradiating the elongated pulse light to an irradiation area; a light receiver disposed at a position to receive transmitted light from a product irradiated with the elongated pulse light, wherein the product moves through the irradiation area without stopping; as well as The judgment unit processes the output data from the light receiver to judge whether the product is good or not. The stretching element stretches the pulses of the light of each wavelength into which the pulsed light is split by the arrayed waveguide diffraction grating using the respective optical fibers. The optical fiber has different lengths depending on the wavelength. The judgment unit is a unit that calculates the content ratio or content of a specific component of the product based on the output data from the light receiver, and compares the calculated content ratio or content with a reference value to thereby judge whether the product is good or not.

9. The product inspection device according to claim 8, characterized in that: The determination unit is a unit that determines whether the output data from the optical receiver is good or not by comparing the value at a predetermined time point in one pulse with a reference value without calculating the light characteristics.

10. The product inspection device according to claim 8, characterized in that: The determination unit is a unit that determines whether the optical characteristic of a predetermined wavelength calculated based on a value at a predetermined time in one pulse of output data from the optical receiver or the value at a predetermined time is good or bad by comparing it with a reference value.

11. The product inspection device according to any one of claims 8 to 10, characterized in that: The pulse light source is a light source that emits pulse light, and the pulse light is light having a continuous spectrum in a wavelength band of at least 1100 nm to 1200 nm.

12. The product inspection device according to any one of claims 8 to 10, characterized in that: The pulse light source is a light source that emits pulse light, and the pulse light is light having a continuous spectrum in a wavelength band of at least 1000 nm to 1300 nm.

13. The product inspection device according to any one of claims 8 to 10, characterized in that: The stretching element is an element that stretches the pulse in a state where the slope of the change in time with respect to the wavelength is 10 picoseconds or more per 1 nm.

14. The product inspection device according to any one of claims 8 to 10, characterized in that: An exclusion mechanism is provided to exclude products judged to be defective.

15. The product inspection device according to any one of claims 8 to 10, characterized in that: The product is a tablet, and the irradiation optical system irradiates the tablet with pulse light having a lengthened pulse from one side in a pattern having a size equal to or larger than that of the tablet.

Images