Method for inspecting foreign matter on sheet surface

Abstract

To provide a method of inspecting a sheet surface for foreign matter, with which it is possible to suppress detection omissions and detect foreign matter.SOLUTION: Provided is a method of inspecting a sheet surface for foreign matter, the sheet being such that when irradiated with light at an angle where the difference between the imaging angle of an imaging device against the sheet surface and the angle of irradiation light against the sheet surface is 40 degrees and an image of the sheet surface is captured by the imaging device, the full width at half maximum of luminance in terms of 256 gray levels in the histogram of the image of the sheet surface is 70 or greater, and the value of peak-top luminance is 80 to 175. This inspection method includes: a step for irradiating light at an angle such that the relative angle of the imaging angle of the imaging device against the sheet surface to the reflection angle of light against the sheet surface, with which the sheet is irradiated, is 0 degrees to 20 degrees, inclusive, and taking a picture with the imaging device to obtain a first image; and a step for irradiating light at an angle such that the relative angle of the imaging angle of the imaging device against the sheet surface to the reflection angle of light against the sheet surface, with which the sheet is irradiated, is 50 degrees to 89 degrees, inclusive, and taking a picture with the imaging device to obtain a second image.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for inspecting foreign matter on the surface of a sheet.

Background Art

[0002] In the process of manufacturing a sheet material, the entry of environmental foreign matter or lumps that appear in the molding process onto the sheet is a problem from the viewpoint of the appearance of the product, and it is common to exclude them as defective products in the inspection process. Foreign matter inspection can be performed visually, but from the viewpoint of improving production efficiency by automation, etc., a system using an inspection device that combines lighting and a photographing device has been studied.

[0003] For example, in Patent Document 1, as a method for inspecting a sheet, a sheet irradiated with light of different brightnesses at a predetermined angle is photographed, and the sheet is inspected by evaluating its luminance. On the other hand, in a heat conduction sheet produced by a method as described in Patent Document 2, most are formed of graphite, and furthermore, color unevenness may easily occur in the manufacturing process. Since there is no difference between the color of the foreign matter and the color of the sheet itself, there has been a background that it is difficult to identify the foreign matter. Specifically, even when the method of Patent Document 1 is used to inspect the sheet of Patent Document 2, many detection failures occur where the foreign matter is hidden in the sheet and cannot be recognized.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the inspection of foreign matters on a sheet such as a heat-conducting sheet mostly formed of graphite, where there is no difference in luminance between the sheet itself and the foreign matters, there was room for further improvement in suppressing detection omissions and detecting foreign matters. Therefore, an object of the present invention is to detect foreign matters without detection omissions in a sheet where there is no difference in luminance between the sheet itself and the foreign matters.

Means for Solving the Problems

[0006] The present inventor conducted intensive studies to achieve the above object. And the present inventor found that it is possible to capture black foreign matters under irradiation conditions where the relative angle between the illumination and the imaging device with respect to the sheet to be inspected is a small angle, it is possible to capture white foreign matters under irradiation conditions where the relative angle between the illumination and the imaging device with respect to the sheet to be inspected is a large angle, and by detecting foreign matters by performing both irradiation conditions step by step, it is possible to suppress detection omissions and detect foreign matters, and thus completed the present invention.

[0007] That is, this invention aims to advantageously solve the above problems, and the method for inspecting foreign matters on the surface of the sheet of the present invention is when the sheet is irradiated with light at an angle where the difference between the imaging angle of the imaging device with respect to the sheet surface and the irradiation angle of light with respect to the sheet surface is 40 degrees, and the image of the sheet surface is captured by the imaging device, the full width at half maximum of the luminance in the histogram of the 256 - tone converted luminance of the image of the sheet surface is 70 or more and the value of the luminance taking the peak top is 80 to 175, the method is a step of irradiating light at an angle where the relative angle between the imaging angle of the imaging device with respect to the sheet surface and the reflection angle of the light irradiated on the sheet with respect to the sheet surface is 0 degrees or more and 20 degrees or less, and capturing an image with the imaging device to obtain a first image; a step of irradiating light at an angle where the relative angle between the imaging angle of the imaging device with respect to the sheet surface and the reflection angle of the light irradiated on the sheet with respect to the sheet surface is 50 degrees or more and 89 degrees or less, and capturing an image with the imaging device to obtain a second image It is an inspection method including: By the irradiation condition at an angle (small angle) where the relative angle is 0 degrees or more and 20 degrees or less, it becomes possible to capture black foreign matters; by the irradiation condition at an angle (large angle) where the relative angle is 50 degrees or more and 89 degrees or less, it becomes possible to capture white foreign matters; by combining both irradiation conditions, it becomes possible to suppress detection omission and detect foreign matters.

[0008] Here, in the method of the present invention, it is preferable that the imaging angle of the imaging device is 90 degrees with respect to the sheet surface. When the imaging angle is 90 degrees (perpendicular), it is possible to suppress unevenness of light irradiation within the field of view of the imaging device and reduce the overlooking rate of foreign matters.

[0009] Further, the method of the present invention preferably further includes a step of calculating, as feature amounts, the brightnesses of different types of foreign matters from the first and second images respectively. By calculating the brightness of foreign matters as a feature amount in this way, it becomes possible to identify the presence of foreign matters based on the obtained digital image data.

Advantages of the Invention

[0010] According to the present invention, it is possible to provide a method for inspecting foreign matters on the sheet surface that can suppress detection omission and detect foreign matters.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Embodiments for Carrying Out the Invention

[0012] Hereinafter, embodiments of the present invention will be described in detail.

[0013] (Foreign object inspection method for sheet surface) The method of the present invention is When irradiating light at an angle where the imaging angle of the imaging device with respect to the sheet surface and the irradiation angle of light with respect to the sheet surface have a difference of 40 degrees, in a sheet where the full width at half maximum of the luminance in the histogram of the 256 - gradation converted luminance of the sheet is 70 or more and the luminance value taking the peak top is 80 - 175, A method of irradiating light on the sheet surface and inspecting for foreign objects on the sheet surface based on an image obtained by imaging the image of the sheet surface with an imaging device, A step (first inspection step) of irradiating light at an angle where the relative angle (first relative angle) between the imaging angle of the imaging device with respect to the sheet surface and the reflection angle of the light irradiated on the sheet is 0 degrees or more and 20 degrees or less, and obtaining a first image by imaging with the imaging device; A step (second inspection step) of irradiating light at an angle where the relative angle (second relative angle) between the imaging angle of the imaging device with respect to the sheet surface and the reflection angle of the light irradiated on the sheet is 50 degrees or more and 89 degrees or less, and obtaining a second image by imaging with the imaging device and including Optionally, further including a step (detection step) of analyzing the first image and the second image to detect foreign objects, which is an inspection method.

[0014] <Applicable sheets> The method of the present invention irradiates light at an angle where the imaging angle of the imaging device with respect to the sheet surface and the irradiation angle of light with respect to the sheet surface have a difference of 40 degrees, and when imaging the image of the sheet surface with the imaging device, the full width at half maximum of the luminance in the histogram of the 256 - gradation converted luminance of the image of the sheet surface is 70 or more and the value of the luminance taking the peak top is 80 - 175. The sheet applicable to the method of the present invention only needs to have at least the inspection surface having the above physical properties, and one side may have the above physical properties, or both sides may have the above physical properties. As will be described later, based on the degree of light and dark of the foreign matter appearance, it is classified into "black foreign matter" and "white foreign matter". In the sheet having the above physical properties, with only one light irradiation and imaging, the luminance difference from either one of the "black foreign matter" and "white foreign matter" becomes small, and it becomes difficult to detect both foreign matters. However, in the method of the present invention, by performing two types of light irradiation and imaging, it becomes easy to detect both the "black foreign matter" and the "white foreign matter". Therefore, the method of the present invention enables effective foreign matter detection for the sheet having the above physical properties. The image acquisition conditions such as the irradiation angle and the imaging angle are as described in the first and second inspection steps described later.

[0015] Note that throughout this specification, the value of "luminance" is represented by the 256 - gradation converted luminance, and refers to the value when the captured image is normalized data in 256 gradations according to the light intensity.

[0016] The full width at half maximum of the luminance is obtained as follows. Let the luminance of the sheet be x, and the luminance histogram be represented as a function f(x). Let the maximum value of f(x) be f max = f(x max ). Assuming f(x) = x max / 2, among the x values that satisfy this, let the minimum x be x1 and the maximum x be x2. Then the full width at half maximum is obtained by the following formula. Full width at half maximum = x2 - x1

[0017] The method of the present invention is preferably applicable to a sheet having a gloss-like appearance (e.g., silver-like gloss, gold-like gloss). Further, the method of the present invention is particularly applicable to a heat conductive sheet, preferably a flame retardant heat conductive sheet, more preferably a heat conductive sheet containing a flame retardant resin and a particulate carbon material. Examples of the sheet to which the method of the present invention is applicable include those described in JP-A-2019-112568 (Patent Document 2).

[0018] <Type of foreign matter> Examples of the "foreign matter" inspected by the method of the present invention include deposits on the sheet surface (e.g., chips, dust, sand, metal foreign matter), precipitates (e.g., graphite, carbon nanotubes (CNT)), altered substances (e.g., oxides), and fine deformations (e.g., unevenness, cracks, streaks, tears). Foreign matter is classified into "black foreign matter" and "white foreign matter" based on the degree of luminance during light irradiation (i.e., the degree of light and dark of the foreign matter appearance). "Black foreign matter" refers to foreign matter detected as a low-luminance foreign matter presenting a dark appearance during light irradiation. Examples of black foreign matter include chips, black spots derived from CNT, and metal foreign matter. "White foreign matter" refers to foreign matter detected as a high-luminance foreign matter presenting a bright appearance during light irradiation. Examples of white foreign matter include dust, sand, fibers such as clothing, and foreign matter derived from the human body such as skin. However, even foreign matter derived from the same substance or state can become black foreign matter or white foreign matter depending on the foreign matter or the surrounding situation.

[0019] <First inspection step> The first inspection step is a step of arranging a light source and an imaging device at a "small relative angle" starting from the sheet surface to obtain an inspection image (first image) of the sheet surface. By arranging the light source and the imaging device at a small relative angle, the luminance difference between the sheet surface and white foreign matter becomes small, while the luminance difference between the sheet surface and black foreign matter becomes large, so that it becomes easy to detect black foreign matter on the sheet surface. In this step, while irradiating the surface of the sheet 111 with the first irradiation light 201a emitted from the first light source 101a, the surface of the sheet 111 is photographed with the imaging device 102 to obtain a first image. Hereinafter, the arrangement relationship among the first light source 101a, the imaging device 102, and the sheet 111 will be described with reference to FIG. 1.

[0020] In FIG. 1, the imaging position 204 is shown for convenience in explaining the arrangement relationship such as the angle. The imaging position 204 refers to the intersection point of the imaging optical axis 203 of the imaging device 102 and the surface of the sheet 111.

[0021] In this specification, the "angle" (e.g., irradiation angle, reflection angle, shooting angle) is expressed as an angle with respect to the surface of the sheet 111 unless otherwise specified.

[0022] The first irradiation angle 211a (α1) refers to the irradiation angle of the first irradiation light 201a. The first irradiation light 201a is reflected at the imaging position 204 on the surface of the sheet 111. The reflected light of the first irradiation light 201a may include diffusely reflected light having various reflection angles. Among the reflected lights of the first irradiation light 201a, the reflected light having a reflection angle equal to the first irradiation angle 211 (α1) (specularly reflected light) is called the first specularly reflected light 202a. Also, the said reflection angle is called the first reflection angle 212a (α1).

[0023] The imaging device 102 captures an image of the surface of the sheet 111 irradiated with the first irradiation light 201a to obtain a first image. The angle of the imaging optical axis 203 of the imaging device 102 (counted from the reflected light side) is called the shooting angle 213 (β). The shooting angle 213 (β) is not particularly limited, but from the viewpoint of being able to irradiate light from a plurality of directions under the same conditions to reduce the miss rate of foreign matters, it is preferably 90 degrees (vertical).

[0024] The relative angle between the shooting angle 213 (β) and the first reflection angle 212a (α1) is called the first relative angle 214a (θ1). The first light source 101a and the imaging device 102 are arranged such that the first relative angle 214a (θ1) is a small angle. In the first inspection step, by making the first relative angle 214a (θ1) a small angle, the detection of black foreign matters becomes easy. The first relative angle 214a (θ1) is 0 degrees or more from the viewpoint of creating a difference between the luminance of the sheet texture and the luminance of the foreign matter, preferably 5 degrees or more, more preferably 10 degrees or more. Also, the first relative angle 214a (θ1) is preferably 20 degrees or less from the same viewpoint.

[0025] <Second inspection process> The second inspection process is a process of arranging a light source and an imaging device at a "large relative angle" starting from the sheet surface to obtain an inspection image (second image) of the sheet surface. By arranging the light source and the imaging device at a large relative angle, while the luminance difference between the sheet surface and black foreign matter becomes smaller, the luminance difference between the sheet surface and white foreign matter becomes larger, so that it becomes easier to detect white foreign matter on the sheet surface. In this process, while irradiating the surface of the sheet 111 with the second irradiation light 201b emitted from the second light source 101b, the surface of the sheet 111 is photographed by the imaging device 102 to obtain a second image. Hereinafter, the arrangement relationship among the second light source 101b, the imaging device 102, and the sheet 111 will be described with reference to FIG. 1.

[0026] The second irradiation angle 211b (α2) refers to the irradiation angle of the second irradiation light 201b. The second irradiation light 201b is reflected at the imaging position 204 on the surface of the sheet 111. The reflected light of the second irradiation light 201b may include diffusely reflected light having various reflection angles. Among the reflected lights of the second irradiation light 201b, the reflected light having a reflection angle equal to the second irradiation angle 211 (α2) (specularly reflected light) is referred to as the second specularly reflected light 202b. Also, the said reflection angle is called the second reflection angle 212b (α2).

[0027] The imaging device 102 photographs the surface of the sheet 111 irradiated with the second irradiation light 201b to obtain a second image. The angle of the imaging optical axis 203 of the imaging device 102 (counted from the reflected light side) is called the photographing angle 213 (β). The photographing angle 213 (β) is not particularly limited, but is preferably 90 degrees (vertical) from the viewpoint of being able to irradiate light from a plurality of directions under the same conditions to reduce the overlooking rate of foreign matter.

[0028] The relative angle between the imaging angle 213(β) and the second reflection angle 212b(α2) is referred to as the second relative angle 214b(θ2). The second light source 101b and the imaging device 102 are arranged such that the second relative angle 214b(θ2) is a large angle. In the second inspection step, by setting the second relative angle 214b(θ2) to a large angle, it becomes easier to detect white foreign matter. The second relative angle 214b(θ2) is 50 degrees or more, preferably 55 degrees or more, and more preferably 58 degrees or more, from the viewpoint of creating a difference between the luminance of the sheet material and the luminance of the foreign matter. Also, from the same viewpoint, the second relative angle 214b(θ2) is 89 degrees or less, preferably 70 degrees or less, and more preferably 75 degrees or less.

[0029] The order in which the first inspection step and the second inspection step are performed is not particularly limited, and the second inspection step may be performed after the first inspection step, the first inspection step may be performed after the second inspection step, or the first inspection step and the second inspection step may be performed alternately.

[0030] <Light source> The wavelength range of the light emitted by the first light source 101a and the second light source 101b may be the visible light region (360 nm to 830 nm), and for example, it may be red (634 nm) or blue (470 nm). Also, the light emitted by the light source may be single-wavelength light (monochromatic light) or mixed-wavelength light in the visible light region (e.g., white light). The wavelengths of the light emitted by the first light source 101a and the second light source 101b may be the same or different.

[0031] The illuminance of the light emitted by the first light source 101a and the second light source 101b on the surface of the sheet 111 is not particularly limited as long as the object of the present invention is achieved, but for example, it may be in the range of 4,000 lx or more and 87,000 lx or less. The illuminance of the light emitted by the first light source 101a and the second light source 101b on the surface of the sheet 111 may be the same or different.

[0032] The type of the light source is not particularly limited, and examples thereof include a white lamp, a fluorescent lamp, an LED (e.g., a red LED, a green LED, a blue LED, a white LED), and a laser light source.

[0033] The first light source 101a and the second light source 101b may each be a single light source or a plurality of light sources. When the first light source 101a or the second light source 101b is a plurality of light sources, it is preferably arranged so that the relative angles with respect to the imaging device 102 are the same. For example, when the imaging angle with respect to the sheet surface of the imaging device 102 is 90 degrees (vertical), the light sources are preferably arranged symmetrically with the imaging position 204 therebetween. The shapes of these light sources are not particularly limited, and examples thereof include a point light source, a bar light source, and an annular light source. Also, the light source may be a light source in which a plurality of light source units are arranged in an array such as a linear or annular shape.

[0034] The first light source 101a and the second light source 101b may be separate light sources or the same light source. When the first light source 101a and the second light source 101b are the same light source, the light source may be movable or fixed. When the light source is fixed, the imaging device 102 or the sheet 111 may move so that the first relative angle 214a (θ1) and the second relative angle 214b (θ2) are within the above-described ranges.

[0035] <Imaging device> The imaging device 102 may be any imaging device that can recognize the luminance of light in 256 (8-bit) or more gradations and acquire image data. Examples of the imaging device 102 include a visible light color camera and a visible light monochrome camera. As the imaging device 102, an imaging device (e.g., a camera) having an integrated semiconductor imaging element (e.g., CMOS, CCD) is preferable. Examples of such an imaging device include an imaging device having a line imaging element (e.g., a 1-pixel width imaging element) (e.g., a line camera) and an imaging device having an area imaging element (e.g., an area camera). From the viewpoint of precisely obtaining luminance data at a predetermined relative angle, an imaging device having a line imaging element is preferable. When the imaging device 102 has a line imaging element, the first light source 101a, the second light source 101b, and the imaging device 102 form an inspection unit, and the inspection unit preferably images while scanning parallel to the surface of the sheet 111, or images the surface of the sheet 111 while moving the sheet 111 parallel to obtain an image of the surface of the sheet 111.

[0036] <Inspection stage> In the first and second inspection steps, the sheet 111 may be placed on the inspection stage 103. The inspection stage 103 may be movable. Examples of the movable inspection stage include a linear movement inspection stage and a rotational inspection stage.

[0037] <Evaluation of foreign object presence> Based on the first image and the second image, the presence of foreign objects on the surface of the sheet 111 is evaluated. A portion where the luminance extremely decreases in the image can be evaluated as a portion where a black foreign object exists, and a portion where the luminance extremely increases in the image can be evaluated as a portion where a white foreign object exists. The first image is used particularly for evaluating the presence of black foreign objects, but may also be used for evaluating the presence of white foreign objects at the same time. The second image is used particularly for evaluating the presence of white foreign objects, but may also be used for evaluating the presence of black foreign objects at the same time.

[0038] The evaluation of the presence of foreign matter based on an image may be performed based on mathematical analysis of digital image data, or may be performed based on visual judgment of an image displayed on a monitor or the like. When the evaluation of the presence of foreign matter is performed based on mathematical analysis of digital image data, anomaly detection using image processing or machine learning techniques may be performed. By performing image processing, for example, noise removal of the image and highlighting of the presence of foreign matter become possible. Examples of image processing include smoothing processing, edge detection processing, binarization processing, combination of the first and second images, and object detection processing by deep learning. Further, the image may be subjected to inversion processing. Examples of machine learning techniques include CNN (Convolutional Neural Network), specifically, R-CNN (Regions with Convolutional Neural Network) for object detection, YOLO (You Only Look Once), SSD (Single Shot MultiBox Detector), etc. may be used.

[0039] The "smoothing process" is an arithmetic process that converts the luminance value of each pixel in digital image data into a luminance value after smoothing using a smoothing filter. For example, for a luminance value array of an n (pixels) × n (pixels) area (n is an odd number of 3 or more, for example, 3 or 5) centered on the target pixel, using a smoothing filter (an n × n coefficient array), it is performed by calculating the sum of the products of the luminance value of the corresponding array element and the coefficient as the luminance value after smoothing. Examples of smoothing filters include Gaussian filters and spline filters. By performing the smoothing process, noise removal of digital image data becomes possible.

[0040] "Edge detection processing" is an arithmetic process that converts the luminance value of each pixel in digital image data into a luminance value after edge detection processing using an edge detection filter. For example, for a luminance value array of an n (pixels) × n (pixels) area (n is an odd number of 3 or more, for example, 3 or 5) centered on a target pixel, using an edge detection filter consisting of a pair of vertical and horizontal arrays (an n×n coefficient array), it is performed by calculating the sum of the products of the luminance values and coefficients of the corresponding vertical and horizontal array elements as the luminance value after processing. Examples of edge detection filters include Sobel filters, Prewitt filters, Roberts filters, and Laplacian filters. By performing edge detection processing, it becomes possible to emphasize the presence of foreign matter.

[0041] "Binarization processing" is a process of classifying the luminance values of each pixel in digital image data into two binary values with a threshold as the boundary. For example, it is performed by converting pixels with a luminance value less than (or less than or equal to) the threshold into "dark pixels" and pixels with a luminance value greater than (or greater than) the threshold into "bright pixels". In the detection of black foreign matter (for example, based on the first image), "dark pixels" are evaluated as "black foreign matter detection pixels", and the site on the sheet surface corresponding to this pixel can be evaluated as a site where black foreign matter exists. Also, "bright pixels" can be evaluated as "black foreign matter non-detection pixels", and the site on the sheet surface corresponding to this pixel can be evaluated as a site where black foreign matter does not exist. In the detection of white foreign matter (for example, based on the second image), "bright pixels" are evaluated as "white foreign matter detection pixels", and the site on the sheet surface corresponding to this pixel can be evaluated as a site where white foreign matter exists. Also, "dark pixels" can be evaluated as "white foreign matter non-detection pixels", and the site on the sheet surface corresponding to this pixel can be evaluated as a site where white foreign matter does not exist. By performing binarization processing, it becomes possible to more prominently identify the sites where foreign matter exists.

[0042] "Inversion processing" is a process of inverting the light and dark of an image. In the image before binarization processing, the inversion processing is, for example, to change the luminance value of each pixel [Luminance value after processing] = [Upper luminance limit value] - [Luminance value before processing] This is done by converting to. In the image after binarization processing, the inversion processing is performed by inverting "dark pixels" and "bright pixels".

[0043] In the image before binarization processing, the "merging" of the first and second images may be performed, for example, by generating an image by adding the luminance values of the pixels of the first and second images, one of which is subjected to inversion processing, or by superimposing the first and second images, one of which is subjected to inversion processing and one or both of which are colored. The first and second images may have the same processing performed on them, or may have different processing performed on them, but it is preferable that the same processing is performed on them.

[0044] In the image after binarization processing, the "merging" of the first and second images may be performed, for example, by designating "black foreign object detection pixels" and "white foreign object detection pixels" as "foreign object detection pixels", designating pixels that do not correspond to either "black foreign object detection pixels" or "white foreign object detection pixels" as "foreign object non-detection pixels", and generating an image in which "foreign object detection pixels" and "foreign object non-detection pixels" are each displayed in a different color (for example, black and white). Alternatively, the "merging" of the first and second images may be performed by generating an image in which "black foreign object detection pixels", "white foreign object detection pixels", and "foreign object non-detection pixels" are each displayed in a different color (for example, different colors and white).

[0045] "CNN (Convolutional Neural Network)" is a neural network model having an input layer for inputting an image, an intermediate layer for extracting feature amounts, and an output layer for identifying the presence or absence and type of an abnormality and outputting the result. The intermediate layer consists of a convolutional layer for extracting feature amounts based on the input pixel values and a pooling layer for reducing the dimension of the feature amounts obtained by convolution. By performing learning from teacher data having information on abnormalities to be identified in advance and constructing a model for the above neural network model, it becomes possible to detect and identify abnormal locations.

[0046] <Feature Quantity Calculation Step> The evaluation of the presence of foreign matter may be performed by a step of calculating, as feature quantities, the brightnesses of different types of foreign matter from the first and second images, respectively. By calculating the brightness of the foreign matter as a feature quantity in this way, it becomes possible to identify the presence of the foreign matter based on the obtained digital image data. Further, by aggregating in advance the correspondence relationship between the brightness value of each foreign matter presence site in the image (before binarization processing) and the type of foreign matter and creating a database, it becomes possible to specify the type of foreign matter according to the brightness value information obtained from the image (before binarization processing).

[0047] <Embodiment Example of the Method of the Present Invention> An example of an embodiment of the method of the present invention is shown in the schematic diagram of FIG. 2. In FIG. 2, the imaging device 102 is arranged directly above the surface of the sheet 111 in a direction perpendicular to the surface of the sheet 111, and light is irradiated from both sides of the imaging position 204. In this arrangement, the imaging angle is 90 degrees (vertical), and the first relative angle 214a (θ1) and the second relative angle 214b (θ2) are equal to the first irradiation angle and the second irradiation angle, respectively. The first light source 101a and the second light source 101b are arranged symmetrically with the imaging device 102 interposed therebetween, whereby the first irradiation light 201a and the second irradiation light 201b are irradiated symmetrically with the imaging optical axis 203 interposed therebetween. By irradiating light from a plurality of directions, the overlooking rate of foreign matter can be reduced. Examples of the light source that irradiates light from a plurality of directions include point light sources and linear light sources arranged at a plurality of positions, and annular light sources.

[0048] A further example of an embodiment of the method of the present invention is shown in the schematic diagram of FIG. 3. In FIG. 3, the sheet 111 is placed on the inspection stage 103. As the inspection stage 103, a rotary inspection stage is used. On the inspection stage 103, an inspection unit 121 having a pair of first light sources 101a, a pair of second light sources 101b, and an imaging device 102 is arranged. The imaging device 102 is arranged at a shooting angle perpendicular (90 degrees) to the surface of the sheet 111. The pair of first light sources 101a and the pair of second light sources 101b are symmetrically arranged with the imaging device 102 interposed therebetween. As the pair of first light sources 101a and the pair of second light sources 101b, bar light sources arranged in parallel are used. As the imaging device 102, it is preferable to use an imaging device having a linear image sensor parallel to the bar light source. The inspection unit 121 is movable in a direction parallel to the surface of the sheet 111 and perpendicular to the bar light source. The first and second inspection steps are performed by scanning (preferably, twice) the inspection unit 121 on the surface of the sheet 111 while switching the first light source 101a and the second light source 101b to obtain images (first and second images) of the surface of the sheet 111. The inspection stage 103 is rotated to move the sheet 111 from the position of the inspection unit 121, and the surface foreign matter of the sheet 111 is inspected by another inspection method (e.g., X-ray inspection), and the foreign matter may be evaluated in combination with the evaluation result of the inspection method of the present invention.

[0049] (Foreign Matter Inspection Device for Sheet Surface) The foreign matter inspection device for the sheet surface used in the method of the present invention is, for example, equipped with a first light source, a second light source, and an imaging device, wherein the first light source, the second light source, and the imaging device are arranged such that the relative angle between the imaging angle of the imaging device with respect to the surface of the sheet to be inspected and the reflection angle of the light irradiated from the first light source onto the sheet with respect to the sheet surface is 0 degrees or more and 20 degrees or less, and the relative angle between the imaging angle and the reflection angle of the light irradiated from the second light source onto the sheet with respect to the sheet surface is 50 degrees or more and 89 degrees or less. It may be an inspection device.

[0050] In the foreign matter inspection apparatus for the sheet surface used in the method of the present invention, from the viewpoint of being able to irradiate light from a plurality of directions under the same conditions to reduce the overlooking rate of foreign matters, the imaging angle with respect to the placement surface of the sheet to be inspected is preferably 90 degrees (vertical).

[0051] The foreign matter inspection apparatus for the sheet surface used in the method of the present invention may further include an inspection stage for installing the sheet to be inspected. Further, in the foreign matter inspection apparatus for the sheet surface used in the method of the present invention, the first light source, the second light source, and the imaging device may form an inspection unit. The inspection unit may be movable parallel to the surface of the sheet to be inspected.

[0052] The preferred examples of each component of the foreign matter inspection apparatus for the sheet surface used in the method of the present invention and the sheet applicable to the inspection apparatus are as described above.

Example

[0053] Hereinafter, the present invention will be specifically described based on examples, but the present invention is not limited to these examples. In the following description, “%” and “parts” representing amounts are based on mass unless otherwise specified.

[0054] (Production Example 1: Production of Thermal Conductive Sheet) <Preparation of Composition> 70 parts of a thermoplastic fluororesin (manufactured by Daikin Industries, Ltd., trade name “Dail G-101”) that is liquid under normal temperature and pressure as a flame retardant resin, 30 parts of a thermoplastic fluororesin (manufactured by 3M Japan Ltd., trade name “Dynion FC2211”) that is solid under normal temperature and pressure, and 90 parts of expanded graphite (manufactured by Ito Graphite Industry Co., Ltd., trade name “EC300”, volume average particle diameter: 50 μm) as a particulate carbon material were stirred and mixed at a temperature of 150° C. for 20 minutes using a pressure kneader (manufactured by Nippon Spindle). Next, the obtained mixture was put into a crusher (manufactured by Osaka Chemical Co., Ltd., trade name “Wonder Crush Mill D3V-10”) and crushed for 10 seconds to obtain a composition.

[0055] <Formation of Pre-Thermal Conductive Sheet> Next, 50 g of the obtained composition was sandwiched between sandblasted PET films (protective films) with a thickness of 50 μm, and roll gap 700 μm, roll temperature 50 °C, roll linear pressure 50 kg / cm, and roll speed 1 m / min were used for rolling (primary pressing) to obtain a pre-thermal conductive sheet with a thickness of 0.8 mm.

[0056] <Formation of Laminate> Subsequently, the obtained pre-thermal conductive sheet was cut into a size of 150 mm in length × 150 mm in width × 0.8 mm in thickness, and 180 sheets were laminated in the thickness direction of the pre-thermal conductive sheet. Further, pressing (secondary pressing) was performed in the lamination direction at a temperature of 23 °C and a pressure of 0.7 MPa for 90 minutes to obtain a laminate with a height of approximately 150 mm.

[0057] <Formation of Thermal Conductive Sheet> Thereafter, leaving the necessary length for slicing, the entire upper surface of the obtained laminate was pressed with a metal plate, and a pressure of 0.1 MPa was applied in the lamination direction (i.e., from above) to fix the laminate. Note that the sides and back of the laminate were not fixed. At this time, the temperature of the laminate was 25 °C. Next, a cutting blade 10 having the shape shown in FIG. 4 (double-edged, blade angle: 20°, maximum thickness of the blade part: 3.5 mm, material: super steel, Rockwell hardness: 91.5, no silicon processing on the blade surface, total length: 200 mm) was attached to the press part of a servo press machine (manufactured by Discharge Precision Machining Laboratory), and slicing was performed in the lamination direction of the laminate (in other words, in the direction coinciding with the normal of the main surface of the laminated pre-thermal conductive sheet) under the conditions of a slicing speed of 200 mm / second and a slicing width of 100 μm to obtain 30 thermal conductive sheets with a size of 150 mm in length × 150 mm in width × 0.10 mm in thickness. Note that the posture of the cutting blade during slicing was such that the angle α shown in FIG. 4 was 10°, and the extending direction of the blade surface 11 was parallel to the slicing surface 21 of the laminate 20. The obtained thermal conductive sheets were used in Examples 1 to 5 and Comparative Examples 1 and 2.

[0058] (Production Example 2: Production of the thermal conductive sheet used in Example 5) The particulate carbon material of Production Example 1 was changed to 50 parts of expanded graphite (manufactured by Ito Graphite Industry Co., Ltd., trade name "EC100", volume average particle diameter: 190 μm). It was produced in the same manner as in Production Example 1 except for this change.

[0059] (Inspection of foreign matter on the surface of the heat conduction sheet) <Example 1> The heat conduction sheet (150 mm × 150 mm × 0.1 mm) obtained in Production Example 1 was placed on an inspection stage (a 200 mm × 200 mm transparent acrylic plate), and white LED bar illumination (model number "LDL2-275X" manufactured by CCS; color temperature 7,800 K; power consumption 27 W) was irradiated from an irradiation angle of 40 degrees with respect to the perpendicular line to the surface of the heat conduction sheet (50 degrees with respect to the surface of the heat conduction sheet), and imaging was performed with an 8K monochrome CMOS line camera (manufactured by Basler; model number "raL8192-12gm"; 8192 px × 1 px) from directly above (imaging angle of 90 degrees with respect to the surface of the heat conduction sheet, the same applies hereinafter). Imaging was performed while scanning the illumination and the camera parallel to the surface of the heat conduction sheet to obtain an 8-bit monochrome image of the sheet surface. Based on the obtained 8-bit monochrome image, a histogram of the 256-level converted luminance (hereinafter simply referred to as "luminance") of the image on the sheet surface was obtained using image processing software (product name "Image J"). The full width at half maximum of the luminance in the histogram was 75. Also, the value of the luminance taking the peak top in the histogram was 95. From the evaluation of these histograms, it was confirmed that the heat conduction sheet obtained in Production Example 1 was applicable to the method of the present invention.

[0060] As the first inspection step, white bar illumination was irradiated from an irradiation angle of 20 degrees with respect to the perpendicular line to the surface of this heat conduction sheet (70 degrees with respect to the surface of the heat conduction sheet), and for the digital image obtained by imaging with an 8K line camera from directly above, the difference between the luminance of the foreign matter and the luminance of the sheet was measured using image processing software. As a result, the luminance range of the sheet was 85 to 220, the luminance range of the target foreign matter was 10 to 50, the maximum value (absolute value) of the luminance of the target foreign matter was 50, and the luminance difference between the maximum value of the foreign matter and the minimum value of the sheet was 35.

[0061] After that, as a second inspection process, white bar illumination was irradiated from an irradiation angle of 60 degrees with respect to the perpendicular to the surface of the heat conduction sheet (30 degrees with respect to the surface of the heat conduction sheet), and the difference between the luminance of foreign matter and the luminance of the sheet was measured by image processing software for the digital image obtained by imaging with an 8K line camera from directly above. As a result, the luminance range of the sheet was 10 to 70, the luminance range of the target foreign matter was 120 to 255, the maximum value (absolute value) of the luminance of the target foreign matter was 120, and the luminance difference between the maximum value of the foreign matter and the minimum value of the sheet was 50.

[0062] After that, as a third step, after performing image processing such as smoothing processing and edge detection on the digital image obtained in the first step and the digital image obtained in the second step by an image processing unit (original software), the binarized image was combined on a database to grasp the details of the foreign matter.

[0063] <Example 2> The irradiation angle with respect to the perpendicular to the surface of the heat conduction sheet in the first step in Example 1 was set to 0 degrees (90 degrees with respect to the surface of the heat conduction sheet). Otherwise, it was the same as in Example 1.

[0064] <Example 3> The irradiation angle with respect to the perpendicular to the surface of the heat conduction sheet in the second step in Example 1 was set to 50 degrees (40 degrees with respect to the surface of the heat conduction sheet). Otherwise, it was the same as in Example 1.

[0065] <Example 4> The irradiation angle with respect to the perpendicular to the surface of the heat conduction sheet in the second step in Example 1 was set to 89 degrees (1 degree with respect to the surface of the heat conduction sheet). Otherwise, it was the same as in Example 1.

[0066] <Example 5> The same procedure as in Example 1 was performed except that the heat conduction sheet obtained in Production Example 2 was used as the heat conduction sheet. From the evaluation of the luminance histogram (shown in Table 1), it was confirmed that the heat conduction sheet obtained in Production Example 2 was applicable to the method of the present invention.

[0067] <Comparative Example 1> In Example 1, the irradiation angle with respect to the perpendicular to the surface of the heat conduction sheet in the first step was set to 30 degrees (60 degrees with respect to the surface of the heat conduction sheet). Otherwise, it was the same as in Example 1.

[0068] <Comparative Example 2> In Example 1, the irradiation angle with respect to the perpendicular to the surface of the heat conduction sheet in the second step was set to 40 degrees (50 degrees with respect to the surface of the heat conduction sheet). Otherwise, it was the same as in Example 1.

[0069] (Evaluation method) <Omission rate> The consistency between the foreign matter information obtained by inspection and the number of foreign matters counted visually was confirmed. The difference in the number of foreign matters missed by inspection compared to the number of foreign matters visually confirmed was classified according to the following criteria. The ratio of the number of foreign matters that were not determined to be foreign matters in the foreign matter information obtained by inspection to the actual number of foreign matters generated was classified according to the following criteria. 〇: The number of omissions is 0% or more and 3% or less △: The number of omissions is more than 3% and 10% or less ×: The number of omissions is more than 10%

[0070] The results are shown in Table 1.

[0071]

Table 1

[0072] From the results in Table 1, it was shown that under the inspection conditions of Examples 1 to 5, the omission rate of foreign matters by inspection was suppressed, and in particular, under the inspection conditions of Examples 1 to 3 and 5, the omission rate of foreign matters by inspection was significantly suppressed.

Industrial applicability

[0073] According to the present invention, it is possible to provide a method for inspecting foreign matters on the surface of a sheet that can suppress detection leakage and detect foreign matters.

Explanation of symbols

[0074] 10 Cutting edge 11 Cutting surface 20 Laminate 21 Slice surface 30 Heat conduction sheet 101a First light source 101b Second light source 102 Imaging device 103 Inspection stage 111 Sheet 121 Inspection unit 201a First irradiation light 201b Second irradiation light 202a First regular reflection light 202b Second regular reflection light 203 Imaging optical axis 204 Imaging position 211a First irradiation angle 211b Second irradiation angle 212a First reflection angle 212b Second reflection angle 213 Shooting angle 214a First relative angle 214b Second relative angle

Claims

1. A method for inspecting foreign matter on the sheet surface, comprising: when the sheet is irradiated with light at an angle such that the difference between the imaging angle of the imaging device with respect to the sheet surface and the irradiation angle of light with respect to the sheet surface is 40 degrees, and an image of the sheet surface is captured by the imaging device, the full width at half maximum of the luminance in the histogram of the 256 - tone converted luminance of the image of the sheet surface is 70 or more, and the luminance value at the peak top is 80 to 175, the heat - conductive sheet containing graphite; the method comprising: a step of irradiating light at an angle such that the relative angle between the imaging angle of the imaging device with respect to the sheet surface and the reflection angle of the light irradiated on the sheet with respect to the sheet surface is 0 degrees or more and 20 degrees or less, and capturing an image with the imaging device to obtain a first image; a step of irradiating light at an angle such that the relative angle between the imaging angle of the imaging device with respect to the sheet surface and the reflection angle of the light irradiated on the sheet with respect to the sheet surface is 50 degrees or more and 89 degrees or less, and capturing an image with the imaging device to obtain a second image; a step of evaluating a portion where the luminance extremely decreases in the first image as a portion where black foreign matter exists; a step of evaluating a portion where the luminance extremely increases in the second image as a portion where white foreign matter exists and including the inspection method.

2. The inspection method according to claim 1, wherein the imaging angle of the imaging device with respect to the sheet surface is 90 degrees.

3. The inspection method according to claim 1, further comprising a step of calculating, as a feature amount, the luminance of different types of foreign matter from the first and second images respectively.

Images