JP2022124077A - Inspection device, inspection method, program, and recording medium - Google Patents

Abstract

To provide an inspection device, an inspection method, a program, and a recording medium capable of detecting a poor sealing including voids.SOLUTION: An inspection device includes: a light irradiation part 10 for irradiating an inspection object W with light in the near infrared region; a transmission spectrum acquisition part 20 for acquiring the transmission spectrum of the light in the near infrared region transmitted through each of a plurality of positions in the inspection object W for each position; a transmission spectrum separation part 61 for separating the transmission spectrum of a specific wavelength band from the transmission spectrum acquired for each position by the transmission spectrum acquisition part 20; and a transmission spectrum analysis part 62 for analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation part 61.SELECTED DRAWING: Figure 1

Description

TECHNICAL FIELD The present invention relates to an inspection apparatus, an inspection method, a program, and a recording medium, and more particularly to an inspection apparatus, an inspection method, a program, and a recording medium for inspecting sealing defects of an object to be inspected.

2. Description of the Related Art A bag-shaped or tube-shaped package containing a content is sealed at an opening after the content is contained in the package. At that time, the sealed portion of the package may contain a gap, or the contents or the waste thereof may be caught. Products with such defective seals must be rejected as defective products.

An X-ray inspection apparatus is known as an inspection apparatus for inspecting the presence or absence of sealing defects in a product whose contents are wrapped in this type of packaging material and sealed. (See, for example, Patent Document 1). An X-ray inspection apparatus inspects an object to be inspected based on the amount of X-ray transmission that passes through the object to be inspected when the object to be inspected is irradiated with X-rays.

Japanese Patent Application Laid-Open No. 2020-60512

However, since the difference in the amount of X-ray transmission between the seal portion including the void and the seal portion not including the void is very small, the conventional X-ray inspection apparatus disclosed in Patent Document 1 does not allow the sealing portion including the void. There was a problem that defects could not be detected.

SUMMARY OF THE INVENTION The present invention has been made to solve such conventional problems, and an object of the present invention is to provide an inspection apparatus, an inspection method, a program, and a recording medium capable of detecting seal defects including voids. and

In order to solve the above problems, an inspection apparatus according to the present invention includes a light irradiation unit that irradiates an object to be inspected with light in the near-infrared region, and a plurality of positions on the object to be inspected. a transmission spectrum acquisition unit that acquires a transmission spectrum of the scattered light in the near-infrared region for each position; and transmission of a specific wavelength band from the transmission spectrum acquired for each position by the transmission spectrum acquisition unit. The configuration includes a transmission spectrum separation section that separates a spectrum, and a transmission spectrum analysis section that analyzes the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation section. The transmission spectrum in the present invention is obtained even when light in the near-infrared region is transmitted through a small distance of the object to be inspected, and not only transmitted light but also reflected light and scattered light Contains transmission spectrum information.

With this configuration, the inspection apparatus according to the present invention can detect sealing defects including voids in the inspection object by selecting and analyzing a specific wavelength band of the transmission spectrum of the inspection object. In addition, the inspection apparatus according to the present invention uses light in the near-infrared region as inspection light to detect the state of the contents in a container made of a resin or the like that does not transmit visible light, the state of the joints of the container, and the like. can be inspected.

Further, in the inspection apparatus according to the present invention, the transmission spectrum analysis unit frequency-analyzes the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation unit, thereby obtaining the transmission spectrum of the specific wavelength band. a frequency analysis unit that obtains a frequency spectrum of; an integration unit that extracts an arbitrary band of the frequency spectrum obtained by the frequency analysis unit and generates integrated information by integrating the intensity of the frequency spectrum in the extracted band; It may be a configuration including

With this configuration, the inspection apparatus according to the present invention can obtain integrated information due to thin-film interference in the inspected object by calculating the frequency spectrum of the transmission spectrum in a specific wavelength band.

Further, in the inspection apparatus according to the present invention, the transmission spectrum analysis unit associates the integration information generated by the integration unit with each of the positions to form a two-dimensional image, so that the characteristics of the object to be inspected are determined. may be configured to further include an imaging unit that generates a feature extraction image obtained by extracting

With this configuration, the inspection apparatus according to the present invention visually displays the features of the inspection object such as seal defects including voids by converting the integrated information resulting from the thin film interference in the inspection object into a two-dimensional image. be able to.

Further, in the inspection apparatus according to the present invention, the imaging unit may be configured to generate the feature extraction image by combining a plurality of mutually different integrated information.

With this configuration, the inspection apparatus according to the present invention can combine different integrated information to emphasize and detect features such as sealing defects of the inspection object.

Moreover, in the inspection apparatus according to the present invention, the transmission spectrum analysis section may be configured by a GPU.

With this configuration, the inspection apparatus according to the present invention can perform a real-time inspection of an object to be inspected in a production line by executing the processing of the transmission spectrum analysis section by the GPU.

Further, in the inspection apparatus according to the present invention, when the object to be inspected contains a resin material, the specific wavelength band is a wavelength band other than the near-infrared absorption band of the resin material. good too.

With this configuration, the inspection apparatus according to the present invention can detect characteristics such as sealing failure of the inspection object by performing analysis while excluding information from the absorption band of the resin material.

Moreover, in the inspection apparatus according to the present invention, the specific wavelength band may be an absorption band in the near-infrared region of the resin material.

With this configuration, the inspection apparatus according to the present invention can detect features derived from the chemical structure of the inspection object based on information from the absorption band of the resin material.

Further, in the inspection apparatus according to the present invention, the light in the near-infrared region irradiated from the light irradiation unit to the inspection object may be divergent light.

With this configuration, the inspection apparatus according to the present invention can effectively acquire information caused by thin-film interference in the inspection object by irradiating the inspection object with divergent light in the near-infrared region.

Further, an inspection method according to the present invention includes a light irradiation step of irradiating an object to be inspected with light in a near-infrared region; A transmission spectrum acquisition step of acquiring a transmission spectrum of light in a region for each position; and a transmission spectrum of separating a transmission spectrum of a specific wavelength band from the transmission spectrum acquired for each position by the transmission spectrum acquisition step. The configuration includes a separation step, and a transmission spectrum analysis step of analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation step.

Further, in the inspection method according to the present invention, the transmission spectrum analysis step performs frequency analysis on the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation step, thereby obtaining the transmission spectrum of the specific wavelength band. a frequency analysis step of obtaining a frequency spectrum of; an integration step of extracting an arbitrary band of the frequency spectrum obtained by the frequency analysis step and generating integration information by integrating the intensity of the frequency spectrum in the extracted band; It may be a configuration including

Further, in the inspection method according to the present invention, the transmission spectrum analysis step associates the integration information generated in the integration step with each of the positions and converts the integration information into a two-dimensional image, thereby determining the characteristics of the object to be inspected. The configuration may further include an imaging step of generating a feature-extracted image from which is extracted.

Further, in the inspection method according to the present invention, the imaging step may be configured to generate the feature extraction image by combining a plurality of mutually different integrated information.

Further, in the inspection method according to the present invention, the transmission spectrum analysis step may be executed on a GPU.

Further, in the inspection method according to the present invention, when the object to be inspected contains a resin material, the specific wavelength band is a wavelength band other than the near-infrared absorption band of the resin material. good too.

Further, in the inspection method according to the present invention, the specific wavelength band may be an absorption band in the near-infrared region of the resin material.

Further, in the inspection method according to the present invention, the light in the near-infrared region with which the object to be inspected is irradiated in the light irradiation step may be divergent light.

A program according to the present invention is a program for causing a computer to execute the above transmission spectrum separation step and any of the above transmission spectrum analysis steps.

A recording medium according to the present invention is a recording medium on which the above program is recorded in a computer-readable manner.

The present invention provides an inspection device, an inspection method, a program, and a recording medium that can detect sealing defects including voids.

1 is a configuration diagram of an inspection device according to an embodiment of the present invention; FIG. (a) is an image showing a test sample with a seal failure, and (b) is a two-dimensional image of the transmission intensity distribution of the test sample. 4 is a graph showing the transmission spectrum at each position of the test sample; (a) is a graph showing the transmission spectrum of the test sample, (b) is a graph showing the frequency spectrum obtained from the transmission spectrum, and (c) gives the greatest intensity in the frequency spectrum of (b). It is the graph which expanded near the frequency. (a) is a feature extracted image obtained from the 0th Fourier component shown in FIG. 4(c), and (b) is a feature extracted image obtained from the first Fourier component shown in FIG. 4(c). . It is an image of a tube as an object to be inspected. 7 is a diagram showing the result of imaging by extracting the spectral intensity of a specific wavelength from the transmission spectrum of the tube of FIG. 6. FIG. FIG. 4 is a diagram showing a transmission spectrum, a frequency spectrum, and a feature extraction image obtained from an inspection object with no seal failure; FIG. 4 is a diagram showing a transmission spectrum, a frequency spectrum, and a feature extraction image obtained from an inspected object with a seal failure; (a) is a feature extraction image based on the ratio of the integrated information of the 0th Fourier component of the non-absorption band A and the integrated information of the first Fourier component of the non-absorption band B; FIG. 4C is a feature extraction image based on the ratio of the integrated information of the Fourier component and the integrated information of the 0th Fourier component of the absorption band E, and FIG. It is a feature extraction image based on the ratio of the integrated information of the Fourier components, and (d) is based on the ratio of the integrated information of the 0th Fourier component in the non-absorbing band C and the integrated information of the 1st Fourier component in the non-absorbing band C. It is a feature extraction image. FIG. 4 shows a transmission spectrum, a frequency spectrum, and a two-dimensional image obtained when the transmission spectrum is Fourier-analyzed without distinguishing between the absorption band and the non-absorption band. FIG. 2 is a conceptual diagram for explaining information contained in an absorption band and a non-absorption band and their extraction; FIG. 4 is a diagram for explaining changes in transmission spectrum according to the degree of divergence of light irradiated to an object to be inspected, where (a) is a configuration diagram of an optical system and (b) is a transmission spectrum; It is a flow chart which shows processing of an inspection method using an inspection device concerning an embodiment of the present invention.

Hereinafter, embodiments of an inspection apparatus, an inspection method, a program, and a recording medium according to the present invention will be described with reference to the drawings.

As shown in FIG. 1, the inspection apparatus 1 according to the embodiment of the present invention includes a light irradiation unit 10, a transmission spectrum acquisition unit 20, a transport unit 30, an operation unit 40, a display unit 50, and a control device 60. And prepare. The inspection apparatus 1 irradiates an object (article) W to be inspected with light in the near-infrared region (hereinafter also referred to as “near-infrared light”), and the light transmitted or reflected by the object W to be inspected at that time. Based on the transmission spectrum of , various inspections such as presence or absence of contamination of foreign matter, presence or absence of defective seal portions, etc. are performed. Here, the object to be inspected W is, for example, a package or container containing a resin material and having a seal portion. The transmission spectrum in the present invention is obtained even when the near-infrared light is transmitted through the object W to be inspected for a short distance, and not only transmitted light but also reflected light and scattered light are transmitted. Contains spectral information.

The light irradiation unit 10 includes a light source 11, a diffuse reflection plate 12, and a slit 13, and irradiates the inspection object W with light in the near-infrared region or a partial wavelength range thereof. there is Here, the near-infrared region corresponds to a wavelength range of 700 nm to 2500 nm.

The light source 11 is a lamp such as a halogen lamp that emits divergent light in the near-infrared region or a partial wavelength range thereof. The diffuse reflection plate 12 diffusely reflects the diverging near-infrared light emitted from the light source 11 . The slit 13 is for irradiating the object W to be inspected W with the near-infrared light diffusely reflected by the diffuse reflection plate 12, and the width of the slit is variable.

The conveying unit 30 is configured by, for example, a belt conveyer arranged horizontally with respect to the apparatus main body, and includes a conveying belt 31 made of a material that easily transmits near-infrared light. The conveying unit 30 sequentially conveys a plurality of inspected objects W in a predetermined conveying direction (the direction indicated by the arrow X) within a conveying path formed by the conveying belt 31 . The conveying unit 30 drives the conveying belt 31 at a conveying speed preset by rotation of a drive motor under control of a conveying control unit (not shown) when inspecting the object W to be inspected. .

The operation unit 40 is for receiving an operation input by the user, and is configured by a touch panel including a touch sensor for detecting a contact position by a touch operation on an input surface corresponding to the display screen of the display unit 50, for example. . When the user touches the position of a specific item displayed on the display screen with a finger or a stylus, the operation unit 40 recognizes the match between the position detected by the touch sensor on the display screen and the position of the item. Thereby, a signal for executing the function assigned to each item is output to the control device 60 . Alternatively, the operation unit 40 may be configured including an input device such as a keyboard or mouse.

The display unit 50 is composed of a display device such as a liquid crystal display or a CRT, and displays various display contents such as analysis results by a transmission spectrum analysis unit 62 described later based on display control by the control device 60 . Furthermore, the display unit 50 displays operation targets such as buttons, soft keys, pull-down menus, and text boxes for setting various conditions.

The control device 60 is composed of a personal computer including, for example, a CPU, GPU, FPGA, ROM, RAM, HDD, etc., and implements the functions of a transmission spectrum separation section 61 and a transmission spectrum analysis section 62, which will be described later. For example, the control device 60 can configure the transmission spectrum separation section 61 and the transmission spectrum analysis section 62 in software by executing a predetermined program by the CPU or GPU. The above program is pre-stored in the ROM or HDD. Alternatively, the above program may be provided or distributed in an installable or executable form recorded on a computer-readable recording medium such as a compact disc or DVD. Alternatively, the above program may be stored in a computer connected to a network such as the Internet and provided or distributed by downloading via the network.

The transmission spectrum acquisition unit 20 is configured by, for example, a hyperspectral camera, and acquires transmission spectra (or absorption spectrum) is acquired for each position on the object W to be inspected. The absorption spectrum can be obtained, for example, by arithmetically processing the transmission spectrum of the object W to be inspected with reference to the transmission spectrum obtained in the absence of the object W to be inspected.

FIG. 2(a) shows an example of a test sample with a seal failure. FIG. 2(b) is a two-dimensional image of the transmission intensity distribution of the test sample generated based on the transmission spectrum of the test sample acquired by the hyperspectral camera. This two-dimensional image of transmission intensity distribution is obtained by integrating transmission intensities of all wavelengths from about 900 nm to 1700 nm.

FIG. 3 shows transmission spectra at positions A to F shown in FIG. 2(b). At positions C, D, and E, no sealing is performed, and a three-layer structure of resin material--gap--resin material is formed. For example, voids may consist of air, product contents, or both. As shown in FIG. 3, at unsealed multi-layered positions C, D, and E, it was found that fringes reflecting thin-film interference due to multi-layering were superimposed near the 1600 nm wavelength band. From the above, it is considered that the presence or absence of fringes in the transmission spectrum can be used to differentiate the sealed and non-sealed portions for imaging. This fringe occurs even in the visible range if the film thickness of the test sample is sufficiently thin, but it is difficult to detect because the amplitude of the fringe is small and the period is short.

The transmission spectrum separating unit 61 separates (cuts out) a transmission spectrum of a specific wavelength band from the transmission spectrum acquired by the transmission spectrum acquisition unit 20 for each position on the object W to be inspected. This specific wavelength band can be arbitrarily set by the user's operation input to the operation unit 40 . For example, the specific wavelength band may be a wavelength band (hereinafter also referred to as a “non-absorption band”) other than the absorption band in the near-infrared region of the material that constitutes the inspection object W. It may be an absorption band in the near-infrared region of the constituent material.

FIG. 4(a) shows the transmission spectrum at position C shown in FIG. For example, the transmission spectrum separation unit 61 extracts the transmission spectrum of all pixels of the two-dimensional image of the transmission intensity distribution shown in FIG. ) is separated from the transmission spectrum of position C.

Transmission spectrum analysis section 62 analyzes the transmission spectrum of the specific wavelength band separated by transmission spectrum separation section 61 , and includes frequency analysis section 63 , integration section 64 , and imaging section 65 . The transmission spectrum analysis unit 62 is configured by a GPU, for example.

The frequency analysis unit 63 performs frequency analysis on the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation unit 61 to obtain the frequency spectrum of the transmission spectrum of the specific wavelength band. Methods of frequency analysis performed by the frequency analysis unit 63 include, for example, discrete Fourier transform, fast Fourier transform, discrete cosine transform, and wavelet transform. A case where the frequency analysis unit 63 performs a discrete Fourier transform or a fast Fourier transform will be described below as an example.

For example, the frequency analysis unit 63 Fourier-transforms the transmission spectrum of the wavelength band enclosed by the dashed line in FIG. Calculate the frequency spectrum as shown. However, data interpolation is performed as preprocessing for Fourier transformation. FIG. 4(c) is an enlarged graph of the vicinity of the frequency giving the highest intensity in the frequency spectrum of FIG. 4(b). Since the fringe component is considered to be the lowest frequency modulation in the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation unit 61, the center frequency component (hereinafter also referred to as "0th Fourier component" ) correspond to the first Fourier components b, b' on both sides of a.

The integrating section 64 extracts an arbitrary band of the frequency spectrum obtained by the frequency analyzing section 63 and generates integration information by integrating the intensity of the frequency spectrum in the extracted band. For example, the integrator 64 calculates the intensity of the first Fourier component b (or b') shown in FIG. ) is output as integration information. Further, the integrating section 64 may output the result of integrating the 0th Fourier component, the second Fourier component, the third Fourier component, and the like as integration information.

The imaging unit 65 associates the integrated information generated by the integrating unit 64 with each position in the object to be inspected W to generate a two-dimensional image, thereby generating a feature extraction image in which the features of the object to be inspected W are extracted. It is designed to FIG. 5(a) is a feature extraction image obtained from the 0th Fourier component a shown in FIG. 4(c). FIG. 5(b) is a feature extraction image obtained from the first Fourier component b shown in FIG. 4(c). From these figures, it was demonstrated that in the feature extraction image obtained from the first Fourier component b, which is considered to reflect the influence of fringes, the sealed portion and the non-sealed portion can be clearly distinguished.

In the transmission spectra shown in FIGS. 3 and 4(a), absorption bands due to molecular vibration of the polymer can be confirmed in the 1200, 1400 and 1700 nm wavelength bands. On the other hand, the fringes of interest in the present invention appear in the non-absorption band. That is, the feature extraction image in FIG. 5(b) focuses on wavelength bands other than the absorption band that is focused in normal spectrum analysis, so that the information derived from the chemical composition of the substance is not reflected or scattered. It shows the result of imaging by extracting information derived from the "shape and structure of the object W to be inspected" reflected in the interference based on the. Such a concept itself is not particularly noteworthy in the field of optical measurement technology involving reference waves such as interferometry, holographic microscopy, and optical tomography. Interference fringe analysis based on light-only irradiation has the potential to develop into a new measurement technology that combines simplicity and robustness.

As described above, it was found that seal failure can be detected by combining a hyperspectral camera and signal processing. Therefore, we decided to verify the effectiveness of the method of the present invention by using a tube, which is a more practical facial cleanser packaging material, as the object W to be inspected, and by further expanding the transmission spectrum region to be analyzed.

First, as an object to be inspected W, a tube T1 as a container for facial cleanser as shown in FIG. 6 was selected as a sample. The sample tube T1 (hereinafter referred to as “tube T1”) is made of a resin film with a thickness of about 0.5 mm, and has a transmittance of about 50% in the near-infrared region. are considered to contain a good balance. A paper label La printed with the product name, ingredients, etc. is attached to the tube T1. This tube T1 is obtained by removing the facial cleanser from the product, washing the inside with water, and drying it. Then, two-dimensional transmission spectrum measurement was performed on the tube T1 using the inspection apparatus 1 of the present embodiment. At this time, the slit width of the slit 13 was about 3 mm, and the conveying speed of the conveying section 30 was 9.1 mm/sec.

FIG. 7 shows the result of imaging with 11 single wavelengths by extracting spectral intensities of specific wavelengths from the transmission spectrum of tube T1. It should be noted that the transmission spectrum in the figure is at one position of the tube T1. At first glance, imaging with a short wavelength emphasizes visually perceptible information such as the label La, while imaging with a long wavelength generally increases the transmittance and reveals the overall characteristics of the tube T1. I know there is. However, even if the transmission spectrum is resolved for each wavelength and imaged, it is not possible to select and extract the features of each portion of the tube T1. Therefore, in this state, it is difficult to perform imaging specialized for inspection items such as seal failure.

In order to solve this problem, the inspection apparatus 1 of the present embodiment performs spectrum analysis by distinguishing between the absorption band and the non-absorption band of the transmission spectrum. That is, the transmission spectrum separating unit 61 separates the transmission spectrum of the absorption band and the transmission spectrum of the non-absorption band from each transmission spectrum acquired by the transmission spectrum acquiring unit 20 for each position on the object W to be inspected.

The upper part of FIG. 8 is a transmission spectrum at one position of the tube T1 where no sealing failure has occurred. Areas A, B, and C surrounded by broken lines indicate non-absorption bands, and areas D and E surrounded by broken lines indicate absorption bands. The middle part of FIG. 8 shows the frequency spectrum obtained by the frequency analysis section 63 from the transmission spectrum of each wavelength band A to E. FIG. In addition, the lower part of FIG. 8 is obtained from the integrated information of the 0th Fourier component a, the integrated information of the first Fourier component b, and the integrated information of the second Fourier component c of the frequency spectrum of each wavelength band A to E. 4 shows a feature extraction image.

The feature extraction image shown in the lower part of FIG. 8 indicates that the information of the tube T1 can be selectively extracted. For example, in the feature extraction image of the 0th Fourier component a of the absorption band D and the feature extraction image of the 0th Fourier component a of the non-absorption band B, the sealing portion at the end of the tube indicated by the dashed ellipse is emphasized for imaging. It is Further, a feature extraction image of the first Fourier component b of the non-absorption band A, a feature extraction image of the first Fourier component b of the non-absorption band B, and a feature extraction image of the first Fourier component b of the non-absorption band C From the comparison, it was observed that the label La of the tube T1 surrounded by the dashed ellipse disappeared as the wavelength lengthened, and the influence of the label La was canceled out.

In order to further verify the above feature extraction, the above tube T1 is cut in half, a plastic case is placed inside, and the seal part is partially cut to create a tube T2 with a "seal failure" added. Created. FIG. 9 shows the results of analyzing the tube T2 as the object W to be inspected under the same procedure and conditions as in FIG. First, in the feature-extracted image of the first Fourier component b of the non-absorbing band B and the feature-extracted image of the first Fourier component b of the non-absorbing band C, the seal failure of the tube T2 existing in the area enclosed by the dashed circle. It can be seen that the part is clearly imaged. In addition, in the area enclosed by the dashed circle in the feature extraction image of the 0th Fourier component a of the non-absorption band C, the plastic case hidden under the label La can be clearly imaged. Therefore, in the feature extraction image of the 0th Fourier component a of the non-absorption band C in FIG. ing. Also, in the feature extraction image of the second Fourier component c of the non-absorption band B and the feature extraction image of the second Fourier component c of the non-absorption band C, the plastic case is imaged in the area enclosed by the dashed circle. I know there is.

Both of the above-mentioned "seal failure" and "enclosed plastic case" are features of the shape and structure of the tube T2. It was confirmed that the

Furthermore, by combining the integrated information of each Fourier component obtained by the above analysis, feature extraction with higher selectivity becomes possible. In other words, the imaging unit 65 may combine a plurality of pieces of cumulative information different from each other to generate a feature extraction image.

For example, FIG. 10(a) shows the ratio of the integrated information of the 0th Fourier component a of the non-absorbing band A and the integrated information of the first Fourier component b of the non-absorbing band B to the tube T2 which is the object W to be inspected. It is a feature-extracted image that is two-dimensionally imaged in association with each position inside. The feature extraction image of FIG. 10(a) is obtained by simultaneously extracting the features of the structure such as the break in the tube T2, the plastic case, and the defective sealing portion.

10(b) shows the ratio of the integrated information of the first Fourier component b of the absorption band D and the integrated information of the 0th Fourier component a of the absorption band E, corresponding to each position in the object W to be inspected. It is a feature extraction image that is two-dimensionally imaged with The feature extraction image of FIG. 10(b) is obtained by selecting the entire tube T2.

FIG. 10(c) shows the ratio of the integrated information of the first Fourier component b of the absorption band D and the integrated information of the 0th Fourier component a of the absorption band D corresponding to each position in the object W to be inspected. It is a feature extraction image that is two-dimensionally imaged with The feature extraction image of FIG. 10(c) is obtained by imaging the difference between the portion where the tube T2 is cut to form one layer and the portion where the tube T2 is formed into two layers.

10(d) shows the ratio of the integrated information of the 0th Fourier component a of the non-absorbing band C to the integrated information of the first Fourier component b of the non-absorbing band C at each position in the object W to be inspected. It is a feature extraction image that is two-dimensionally imaged in association with . The feature extraction image of FIG. 10(d) is obtained by selectively extracting the included plastic case by canceling out all the effects of the material of the tube T2 and the label La.

Contrary to the above-described procedure, when the absorption band and the non-absorption band of the transmission spectrum obtained by the transmission spectrum acquisition unit 20 are not distinguished, and the entire area F is collectively subjected to Fourier analysis, the same result as in FIG. FIG. 11 shows the results of the analysis performed under the procedures and conditions. First, there is no clear boundary between the Fourier components in the frequency spectrum of the full F. Therefore, the two-dimensional images respectively obtained from the 0th Fourier component a, the 1st Fourier component b, the 2nd Fourier component c, and the 3rd Fourier component d of the full range F selectively image the features of the tube T2. It is not what was done.

As described above, the inspection apparatus 1 of the present embodiment separates the transmission spectrum of the inspection object W into an absorption band and a non-absorption band and performs Fourier analysis. Basically, the transmission spectrum Information on the shape and structure of the object W to be inspected is extracted by utilizing the periodicity and amplitude of the fringes superimposed on the . On the other hand, the absorption bands contain information based on the chemical details of the materials themselves that make up the object W to be inspected. Conceptually, as shown in FIG. 12, a coloring book can be used as an example. First, the structural and shape information appearing in the non-absorbing zone corresponds to the border of the coloring book. On the one hand, the chemical information that appears in the absorption bands corresponds to the colors in the coloring book.

Therefore, by separately handling the information obtained from the absorption band and the information obtained from the non-absorption band, the information of the object W to be inspected can be selectively extracted and imaged. Based on this understanding, when looking at the two-dimensional image composed of the Fourier components in FIG. 11, it can be seen that the frames and colors of the coloring book are mixed. In particular, in the two-dimensional image of the first Fourier component b and the two-dimensional image of the second Fourier component c, various elements are mixed and imaged, and it looks like an abstract painting.

The features of the light irradiation unit 10 included in the inspection apparatus 1 of this embodiment will be described below. In commercially available spectrophotometers, divergence of the light irradiated onto the inspection object is considerably suppressed by lengthening the optical path length from the lamp light source to the inspection object and shaping the light using a concave mirror. On the other hand, in the light irradiation unit 10 according to the present embodiment, the divergent light emitted from the light source 11 such as a halogen lamp is directly irradiated onto the object W to be inspected through the slit 13, so that the object W is irradiated with the divergent light. The degree of light divergence is extremely high. The effect of the difference in the degree of divergence on the transmission spectrum of the object W to be inspected was verified using an optical system as shown in FIG. 13(a).

Light from a halogen lamp 70 passes through a pinhole 71, is diverged by a lens 72 placed at a focal point, and enters a tube film sample 73 having a thickness of about 0.5 mm. The light emitted from the sample 73 enters a Fabry-Perot type spectroscope 74, and the transmission spectrum of the sample 73 is obtained. The smaller the diameter of the pinhole 71 , the closer it becomes to a minute point light source based on Huygens' principle. On the other hand, the larger the diameter of the pinhole 71, the more divergent the light passing through the lens 72 becomes. As shown in FIG. 13B, the more the light passing through the lens 72 becomes divergent, the more fringes are superimposed on the light after passing through the sample 73 .

The origin of fringing in the diverging light that has passed through this thick sample 73 is considered as follows. As a precondition, interference occurs when light beams with optical path differences overlap each other like Newton's rings. Therefore, when parallel light is incident on the sample 73, no interference fringes occur (however, film is thin). Moreover, even if the light is divergent, no interference fringes are visible unless it is incident on the sample 73 (there is no fringe in the transmission spectrum of the light from the lamp). It can also be said that this is because interference occurs everywhere and interference fringes of a specific wavelength do not appear.

On the other hand, if the diverging light is made incident on the thick sample 73, the optical path length of the light propagating through the sample 73 changes depending on the incident angle. This reduces the spatial coherency at certain wavelengths of light that was interfering everywhere, while leaving (or improving) the coherence at other wavelengths, leaving the relative interference fringes emerges. As a result, it is believed that the fringes will be slightly superimposed on the transmission spectrum of the thick sample 73 upon which the divergent light is incident.

It is clear that the generation and enhancement of fringes by divergent light play an important role in the inspection technique for extracting the features of the object W according to the present invention. It is also advantageous in that it does not require

An example of processing of an inspection method using the inspection apparatus 1 according to the present embodiment will be described below with reference to the flowchart of FIG. 14 .

First, the control device 60 controls the range of the absorption band and the non-absorption band of the object to be inspected W, the combination of the wavelength band and the Fourier component used when generating the feature extraction image, according to the operation of the operation unit 40 by the user. and other information are set (step S1).

Next, the transport unit 30 starts transporting the inspection object W (transport step S2).

Next, the light irradiation unit 10 irradiates the inspection object W with near-infrared light (light irradiation step S3).

Next, the transmission spectrum acquisition unit 20 acquires the transmission spectrum of the near-infrared light transmitted, reflected, or scattered at each of the plurality of positions on the object W to be inspected (transmission spectrum Acquisition step S4).

Next, the transmission spectrum separating unit 61 separates the transmission spectrum of a specific wavelength band (one or more) from the transmission spectrum acquired for each position on the object to be inspected W in the transmission spectrum acquisition step S4 (transmission spectrum separation step S5).

Next, the frequency analysis unit 63 obtains the frequency spectrum of the transmission spectrum of the specific wavelength band by frequency-analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation step S5 (frequency analysis step S6). .

Next, the integration unit 64 extracts arbitrary bands such as the 0th Fourier component, the 1st Fourier component, and the 2nd Fourier component of the frequency spectrum obtained in the frequency analysis step S6, and the intensity of the frequency spectrum in the extracted band is generated by integrating (accumulation step S7).

Next, the control device 60 determines whether or not integration information has been obtained in the integration step S7 for all desired wavelength bands separated in the transmission spectrum separation step S5 (step S8). When the integration information is acquired in the integration step S7 for all wavelength bands, the control device 60 executes the process of step S9. On the other hand, if the integration information has not been acquired by the integration step S7 for all wavelength bands, the control device 60 executes the processing from step S6 onward again.

In step S9, the imaging unit 65 associates the accumulated information generated in the accumulation step S7 with each position on the object W to be two-dimensionally imaged, thereby extracting the characteristics of the object W to be a feature extracted image. is generated (imaging step S9). For example, the imaging unit 65 generates a feature extraction image for each wavelength band based on the integrated information for each wavelength band. Alternatively, the imaging unit 65 generates a feature extraction image by combining integrated information of a plurality of mutually different wavelength bands. Alternatively, the imaging unit 65 generates a feature extraction image by combining a plurality of integrated information for a certain wavelength band.

Next, the display unit 50 displays the feature extraction image generated in the imaging step S9 (step S10).

Note that the above steps S6 to S9 constitute a transmission spectrum analysis step which is executed on, for example, the GPU and analyzes the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation step S5.

As described above, the inspection apparatus 1 according to the present embodiment selects and analyzes a specific wavelength band of the transmission spectrum of the inspection object W, thereby detecting sealing defects including voids in the inspection object W. be able to. In particular, the inspection apparatus 1 can inspect the state of contents in a container made of resin or the like through which visible light does not pass, the state of joints of the container, and the like, by using near-infrared light as the inspection light. can. In this embodiment, the process of analyzing the near-infrared light transmitted through the object W to be inspected has been described as an example, but the present invention is not limited to this. For example, for an object to be inspected W having a relatively low transmittance, the same analysis can be performed by using both the reflected light and the transmitted light from the object to be inspected W, or by using only the reflected light.

Further, the inspection apparatus 1 according to the present embodiment can obtain integrated information due to thin-film interference in the inspection object W by calculating the frequency spectrum of the transmission spectrum in a specific wavelength band.

In addition, the inspection apparatus 1 according to the present embodiment visualizes the characteristics of the object W to be inspected, such as sealing defects including voids, by forming a two-dimensional image of integrated information resulting from thin film interference in the object W to be inspected. can be displayed.

In addition, the inspection apparatus 1 according to the present embodiment can emphasize and detect the characteristics of the inspection object W, such as a sealing failure, by combining different accumulated information.

Moreover, the inspection apparatus 1 according to the present embodiment can perform real-time inspection of the inspection object W in the production line by executing the processing of the transmission spectrum analysis unit 62 by the GPU.

In addition, the inspection apparatus 1 according to the present embodiment can detect characteristics such as a sealing failure of the inspection object W by performing analysis while excluding information from the absorption band of the resin material.

Moreover, the inspection apparatus 1 according to the present embodiment can detect features derived from the chemical structure of the inspection object W based on information from the absorption band of the resin material.

In addition, the inspection apparatus 1 according to the present embodiment can effectively acquire information due to thin film interference in the inspection object W by irradiating the inspection object W with divergent light in the near-infrared region. .

Reference Signs List 1 inspection device 10 light irradiation unit 11 light source 12 diffuse reflector 13 slit 20 transmission spectrum acquisition unit 30 transport unit 31 transport belt 40 operation unit 50 display unit 60 control device 61 transmission spectrum separation unit 62 transmission spectrum analysis unit 63 frequency analysis unit 64 Integration unit 65 Imaging unit W Object to be inspected

Claims (18)

a light irradiator (10) that irradiates an object (W) to be inspected with light in the near-infrared region;
a transmission spectrum acquisition unit (20) for acquiring, for each position, a transmission spectrum of light in the near-infrared region transmitted, reflected, or scattered at each of a plurality of positions on the object to be inspected;
a transmission spectrum separation unit (61) for separating a transmission spectrum of a specific wavelength band from the transmission spectrum acquired for each position by the transmission spectrum acquisition unit;
and a transmission spectrum analysis section (62) for analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation section. The transmission spectrum analysis unit is
a frequency analysis unit (63) for obtaining a frequency spectrum of the transmission spectrum of the specific wavelength band by frequency-analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation unit;
an integrator (64) that extracts an arbitrary band of the frequency spectrum obtained by the frequency analysis unit and generates integrated information by accumulating the intensity of the frequency spectrum in the extracted band. The inspection device according to claim 1. The transmission spectrum analysis unit is
It further includes an imaging unit (65) for generating a feature extraction image in which features of the inspection object are extracted by associating the integration information generated by the integration unit with each of the positions and forming a two-dimensional image. 3. The inspection apparatus according to claim 2, characterized in that: 4. The inspection apparatus according to claim 3, wherein the imaging unit combines a plurality of pieces of integrated information different from each other to generate the feature extraction image. 5. The inspection apparatus according to any one of claims 1 to 4, wherein said transmission spectrum analysis unit is configured by a GPU. When the object to be inspected contains a resin material,
The inspection apparatus according to any one of claims 1 to 5, wherein the specific wavelength band is a wavelength band other than an absorption band in the near-infrared region of the resin material. 6. The inspection apparatus according to any one of claims 1 to 5, wherein the specific wavelength band is an absorption band in the near-infrared region of the resin material. 8. The inspection apparatus according to any one of claims 1 to 7, wherein the light in the near-infrared region irradiated from the light irradiation unit to the object to be inspected is divergent light. A light irradiation step (S3) of irradiating the object (W) to be inspected with light in the near-infrared region;
a transmission spectrum acquisition step (S4) of acquiring, for each position, a transmission spectrum of the light in the near-infrared region transmitted, reflected, or scattered at each of a plurality of positions on the object to be inspected;
a transmission spectrum separation step (S5) of separating a transmission spectrum of a specific wavelength band from the transmission spectrum acquired for each position in the transmission spectrum acquisition step;
and transmission spectrum analysis steps (S6 to S9) for analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation step. The transmission spectrum analysis step includes:
a frequency analysis step (S6) of obtaining a frequency spectrum of the transmission spectrum of the specific wavelength band by frequency-analyzing the transmission spectrum of the specific wavelength band separated by the transmission spectrum separation step;
and an integration step (S7) of extracting an arbitrary band of the frequency spectrum obtained by the frequency analysis step and generating integration information by integrating the intensity of the frequency spectrum in the extracted band. The inspection method according to claim 9. The transmission spectrum analysis step includes:
further comprising an imaging step (S9) of generating a feature extraction image in which features of the object to be inspected are extracted by associating the integration information generated by the integration step with each of the positions and forming a two-dimensional image. The inspection method according to claim 10, characterized in that: 12. The inspection method according to claim 11, wherein said imaging step combines a plurality of pieces of integrated information different from each other to generate said feature extraction image. 13. The inspection method according to claim 9, wherein said transmission spectrum analysis step is executed on a GPU. When the object to be inspected contains a resin material,
14. The inspection method according to any one of claims 9 to 13, wherein the specific wavelength band is a wavelength band other than an absorption band in the near-infrared region of the resin material. 14. The inspection method according to any one of claims 9 to 13, wherein the specific wavelength band is an absorption band in the near-infrared region of the resin material. 16. The inspection method according to any one of claims 9 to 15, wherein the light in the near-infrared region with which the object to be inspected is irradiated in the light irradiation step is divergent light. A program for causing a computer to execute the transmission spectrum separation step according to claim 9 and the transmission spectrum analysis step according to any one of claims 9 to 12. 18. A recording medium on which the program according to claim 17 is recorded in a computer-readable manner.

Images