WO2024014273A1

WO2024014273A1 - Inspection system, and method for correcting angle of inclination of object surface using same

Info

Publication number: WO2024014273A1
Application number: PCT/JP2023/023635
Authority: WO
Inventors: 秋希良藤井; 一馬原口; 泰資田中; 翔馬高橋; 匠羽根田
Original assignee: パナソニックＩｐマネジメント株式会社
Priority date: 2022-07-13
Filing date: 2023-06-26
Publication date: 2024-01-18
Also published as: TW202407332A

Abstract

This inspection system 100 comprises at least an irradiation device 50 that irradiates an object 200 with inspection light, and a camera (imaging device) 80 that captures an image of the object 200 irradiated with the inspection light. The irradiation device 50 is provided with: a planar light source 10 that emits the inspection light; and a color filter 20 that is disposed between the planar light source 10 and the object 200 in the optical path of the inspection light, the color filter 20 including a plurality of filters that transmit light in mutually different wavelength bands. The inspection device detects inclination of the surface of the object on the basis of a reference image and the image of the object 200 that was captured by the camera 80.

Description

Inspection system and method for correcting the tilt angle of the surface of an article using the same

The present disclosure relates to an inspection system used for visual inspection of an article and a method of correcting the inclination angle of the surface of the article using the same.

Conventionally, configurations disclosed in Patent Documents 1 and 2 are known as illumination devices used for visual inspection of articles.

In the conventional configurations disclosed in Patent Documents 1 to 3, a light shielding filter is placed between the surface light source and the lens. Further, this light shielding filter is arranged at the focal position on the incident side of the lens. By configuring the lighting device in this way, the lighting conditions can be made the same at each point on the surface of the article. Further, as disclosed in Patent Document 3, by combining a light shielding filter and a color filter, the color distribution of reflected light can be changed depending on the direction (inclination) of the surface of the article. As a result, in an inspection system equipped with an illumination device, the direction of the surface of the article, surface irregularities, scratches, etc. can be detected with high precision.

Patent No. 5866573 Patent No. 5866586 Patent No. 6451821

By the way, in an inspection system, inspection light emitted from a lighting device is reflected on the surface of an article, and the reflected light is incident on a camera to capture an image of the article. Generally, images of articles are obtained as color images. In such a case, it is necessary to calibrate the color balance in the captured image.

Here, since the reflectance differs depending on the component of the article (gold, copper, etc.), it is necessary to calibrate the color balance according to the component of the article. In particular, when the color filter includes a color distribution with a gradation in which the color scheme changes continuously, it is difficult to adjust the color balance of the color image. Further, even for the same article, the reflectance may differ depending on the wavelength of the irradiated light. In such a case, it is also difficult to adjust the color balance of a color image.

An object of the present disclosure is to provide an inspection system that can adjust the color balance of an image of an article, and a method of correcting the tilt angle of the surface of an article using the same.

In order to achieve the above object, an inspection system according to the present disclosure includes at least an illumination device that irradiates an article with inspection light, and an imaging device that images the article irradiated with the inspection light. The illumination device includes a light source that emits the inspection light, and a plurality of filters that are disposed between the light source and the article in the optical path of the inspection light and that transmit light having different wavelength widths. A color filter is provided, and the inspection system is characterized in that the image of the article captured by the imaging device is compared with a reference image to detect the inclination of the surface of the article.

A method of correcting a tilt angle of a surface of an article according to the present disclosure is a method of correcting a tilt angle of a surface of an article using the inspection system, in which the article is irradiated with the inspection light and the image capturing device is used to correct the tilt angle of the article. a first step of acquiring an image; a second step of setting a reference plane on the surface of the article; a third step of calculating an offset based on the image of the reference plane; a fourth step of correcting the inclination angle by subtracting the offset calculated in the third step from the inclination angle at each part of the surface of the article; It is characterized by corresponding to the inclination of the article.

According to the present disclosure, the color balance of an image of an article can be adjusted.

FIG. 1A is a schematic configuration diagram of an inspection system according to the first embodiment. FIG. 1B is an enlarged view of the portion surrounded by the broken line in FIG. 1A. FIG. 2 is a schematic diagram showing the solid angle of irradiation of the inspection light irradiated onto the article. FIG. 3 is a schematic plan view of the color filter. FIG. 4 is a diagram showing an example of the wavelength dependence of the reflectance of metal. FIG. 5 is a schematic plan view of a color filter used during article inspection. FIG. 6 is a cross-sectional view of the article used during the article inspection. FIG. 7 is an image captured by a camera during an article inspection. FIG. 8 is a schematic plan view of a color filter according to the second embodiment. FIG. 9 is a diagram showing the color distribution within the light exit plane of the surface light source according to the second embodiment. FIG. 10 is a diagram showing the rotation angle dependence of light brightness at the outermost periphery of a surface light source. FIG. 11 is a schematic diagram illustrating the color distribution of inspection light according to a modification. FIG. 12A is a schematic diagram showing the color distribution of blue light in the inspection light. FIG. 12B is a schematic diagram showing the color distribution of red light in the inspection light. FIG. 13 is a schematic configuration diagram of an inspection system according to the third embodiment. FIG. 14 is a schematic diagram of an image of an article when the article is not correctly installed on the support stand.

Hereinafter, embodiments of the present disclosure will be described based on the drawings. Note that the following description of preferred embodiments is essentially just an example, and is not intended to limit the present disclosure, its applications, or its uses.

(Embodiment 1)
[Configuration and operation of inspection system]
FIG. 1A shows a schematic configuration diagram of an inspection system according to this embodiment, and FIG. 1B shows an enlarged view of a portion surrounded by a broken line in FIG. 1A. FIG. 2 schematically shows the irradiation solid angle of the inspection light irradiated onto the article. In the following description, the optical axis direction of the inspection light may be referred to as the X direction, and the direction from the support base 70 toward the camera 80 may be referred to as the Z direction. The direction that intersects the X direction and the Z direction is sometimes called the Y direction.

As shown in FIG. 1, the inspection system 100 includes an illumination device 50, a half mirror (optical member) 60, a support stand 70, and a camera (imaging device) 80. The illumination device 50 and the half mirror 60 are arranged inside the housing 40.

As will be described in detail later, in the inspection system 100, the article 200 placed on the support stand 70 is illuminated with inspection light from the illumination device 50, and the reflected light reflected by the article 200 is imaged by the camera 80. The appearance of the article 200 is inspected based on the image captured by the camera 80.

The illumination device 50 is composed of a surface light source 10, a color filter 20, and a lens 30, and the surface light source 10 emits white planar light as inspection light.

The color filter 20 is disposed between the surface light source 10 and the lens 30 at the focal position of the lens 30, in this case at the focal position on the incident side. The color filter 20 has a predetermined color distribution within the plane of incidence of the inspection light. Therefore, the inspection light emitted from the surface light source 10 passes through the color filter 20 and becomes plane light having the above-mentioned color distribution in the direction intersecting the traveling direction of the inspection light. The color distribution given to the color filter 20 will be described in detail later.

The lens 30 focuses the inspection light emitted from the surface light source 10 and transmitted through the color filter 20 toward the article 200. Further, the inspection light is given a predetermined irradiation solid angle IS (see FIG. 2) by passing through the lens 30. I will discuss this further.

As shown in FIG. 1, when a surface light source 10, a color filter 20, and a lens 30 are arranged, the irradiation solid angle IS at a point P1, which is on the optical axis of the inspection light and is the exit-side focal position of the lens 30, is is uniquely determined by the diameter of the optical path of the inspection light in the color filter 20 and the focal length f of the lens 30. Note that the "irradiation solid angle" referred to here refers to an arbitrarily shaped cone that has a predetermined point on the optical path of the inspection light as its apex and indicates the range in which light is irradiated to the predetermined point (for example, as shown in Fig. (see 2). When the diameter of the above-mentioned optical path is r, the plane half angle θ of the irradiation solid angle IS satisfies the relationship shown in equation (1).

Furthermore, even if the position is away from the optical axis of the inspection light, the irradiation solid angle IS at a position away from the center of the lens 30 by the focal position on the exit side of the lens 30 has the same shape as the irradiation solid angle IS at the point P1. will be the same size. Further, the irradiation solid angle IS at a position farther from the exit-side focal position of the lens 30 also has the same shape and the same size as the irradiation solid angle IS at the point P1.

Further, even when the inspection light is reflected by the half mirror 60, these irradiation solid angles IS are maintained. Therefore, as shown in FIG. 2, the inspection light is irradiated so that each point on the surface of the article 200 has the same irradiation solid angle IS. That is, at any point on the surface of the article 200, the illumination conditions are the same regardless of the distance from the surface light source 10.

The half mirror (optical member) 60 is placed in the optical path of the inspection light transmitted through the lens 30 and in the optical path of the reflected light reflected by the article 200 toward the camera 80. The half mirror 60 reflects the inspection light focused by the lens 30 toward the article 200, while transmitting the reflected light reflected by the article 200. In other words, the half mirror 60 irradiates the inspection light onto the article 200 and causes the reflected light from the article 200 to enter the camera 80 .

The support stand 70 has a flat surface, and the article 200 is placed on the surface. The article 200 has a region S1 that is a plane parallel to the X direction and the Y direction, a region S2 that is inclined toward the left side of the drawing in FIG. 1B, and a region S3 that is inclined toward the right side of the drawing in FIG. 1B.

The camera (imaging device) 80 receives the reflected light reflected by the article 200 and transmitted through the half mirror 60, and generates an image of the article 200. That is, the camera 80 captures an image of the article 200 irradiated with the inspection light. Note that the image captured by the camera 80 is a color image. Therefore, although not shown, the camera 80 has a CMOS image sensor equipped with a color filter as an image sensor having an imaging surface. The camera 80 may also include a processor (not shown) that processes the output signal of the CMOS image sensor to generate an image.

Next, the operating principle of the inspection system 100 will be explained.

As shown in FIG. 1, the inspection light is reflected by the half mirror 60 so that the optical axis points in the Z direction. When the surface of the article 200 is flat and parallel to the surface of the support base 70, the reflected light reflected from the surface of the article 200 (hereinafter simply referred to as reflected light) becomes white light. For example, the reflected light from area S1 on the surface of article 200 shown in FIG. 1B is incident on camera 80 as white light.

On the other hand, the color filter 20 has a color distribution within the plane of incidence of the inspection light. As shown in FIG. 3, the color filter 20 includes a plurality of filters 20a arranged in an array in the X direction and the Y direction. Each filter 20a is arranged with an interval d between adjacent filters 20a in the X direction and the Y direction.

Further, the plurality of filters 20a include a red filter (R filter) that transmits red light, a green filter (G filter) that transmits green light, and a blue filter (B filter) that transmits blue light. In FIG. 3, the area where the red filter is provided is shown as "R", the area where the green filter is provided as "G", and the area where the blue filter is provided as "B". Specifically, in the color filter 20, the color distribution is R:G:B=7:1:1 in area 201 (the area at the upper left end of the drawing), and the color distribution is R:G:B=7:1:1 in the area 202 (the area at the upper right end of the drawing). is R:G:B=2:7:0, and in area 203 (the area at the bottom left edge of the drawing), the color distribution is R:G:B=2:0:7, and in area 204 (the area at the bottom right edge of the drawing), the color distribution is R:G:B=2:0:7. ), the color distribution is R:G:B=3:3:3.

Note that the R component (red light), the G component (green light), and the B component (blue light) each have a predetermined wavelength width. For example, the R component includes a wavelength of 650 nm and has a wavelength width of approximately several tens of nm to 100 nm. The G component includes a wavelength of 500 nm and has a wavelength width of approximately several tens of nm to 100 nm. The B component includes a wavelength of 450 nm and has a wavelength width of approximately several tens of nm to 100 nm.

As a result, the inspection light according to the color distribution of the color filter 20 is reflected by the surface of the article 200 and enters the imaging surface of the camera 80. Therefore, by comparing the reference image and a captured image captured with the article 200 placed, it is possible to estimate the direction (inclination) of the surface of the article 200. The reference image at this time is an image of the article 200 captured using the color filter 20 in which only the R filter is arranged, the color filter 20 in which only the G filter is arranged, and the color filter 20 in which only the B filter is arranged. be. If a monochromatic filter is used as the color filter 20, even if the surface of the article 200 is tilted, the brightness value will be constant, making it possible to calibrate the equation (3) described below. . Note that images taken of the article 200 (or an article made of the same material as the article 200) from a plurality of known predetermined angles may be used as the reference image. Even in this case, calibration of equation (3) described below is possible. Furthermore, images taken of the article 200 (or an article made of the same material as the article 200) from a plurality of unknown predetermined angles may be used as the reference image. Even in this case, it is possible to calibrate equation (3) described below based on the relationship between the filters. Furthermore, if the article 200 has a plurality of surfaces having different angles on its surface, a single image of the article 200 may be used as the reference image. Even in this case, the following equation (3) can be calibrated based on the angle of the surface of the article 200 and the RGB ratio of the captured image.

For example, when the color filter 20 in FIG. 3 is used, light emitted from the area 201 is focused on one pixel of the camera 80. Therefore, by detecting an area where R:G:B=7:1:1 in the captured image generated by the camera 80, it is possible to detect an area corresponding to the area 201 in the captured image. Note that in FIG. 3, the light emitted from each of the regions 201 to 204 is focused on a corresponding one pixel in the camera 80, but the light emitted from each of the regions 201 to 204 is focused on a corresponding pixel in the camera 80; The light may be focused on one pixel in the camera 80.

Note that the arrangement of the plurality of filters 20a in the color filter 20 is not limited to the example in FIG. 3 (6×6 filters 20a are arranged), and the arrangement of the filters 20a can be set arbitrarily. For example, the color filter 20 may include 9×9 or 12×12 filters 20a. Further, in the example of FIG. 3, a 3×3 filter 20a corresponds to one pixel of the camera 80, but the number and arrangement of filters 20a corresponding to one pixel of the camera 80 can be set arbitrarily. For example, one pixel of the camera 80 may correspond to a 6×6 or 9×9 filter 20a. Further, the color distribution of the plurality of filters 20a in the color filter 20 can be arbitrarily set.

As explained above, according to the inspection system 100 shown in FIG. 1A, the color filter 20 includes a red filter that transmits red light, a green filter that transmits green light, and a blue filter that transmits blue light. included. The red filter, green filter, and blue filter transmit light with different wavelength widths, so by setting the ratio of RGB incident to one pixel of the camera 80 in advance, it is possible to quantify RGB in the captured image. expression becomes possible. That is, it is possible to know from which position of the color filter 20 the light emitted corresponds to which position in the captured image. Therefore, by comparing in advance a reference image with no article 200 placed thereon and a captured image with the article 200 placed thereon, it is possible to estimate the inclination of the surface of the article 200.

Note that as the reference image, a captured image of the article 200 (hereinafter sometimes referred to as a non-defective image) that has been previously determined to have no abnormalities in a visual inspection may be used. By taking the difference between this non-defective product image and the captured image of the product 200 to be inspected (hereinafter sometimes referred to as the product to be inspected), it is possible to determine the presence or absence of abnormalities on the surface of the product to be inspected. can. For example, it is possible to determine the presence or absence of scratches or stains on the surface of an item to be inspected, as well as the location of any scratches or stains, if any. Furthermore, if there is a location on the surface of the item to be inspected whose angle of inclination is out of the allowable range, the position of the location and the angle of inclination can be estimated. In addition, in taking the difference between the non-defective product image and the captured image of the product to be inspected, a method similar to the procedure for taking the difference between general images can be adopted. For example, the difference may be taken after matching the imaged position and orientation of the article 200 between the non-defective product image and the captured image of the inspection target.

Further, the plurality of filters 20a are arranged in the color filter 20 such that adjacent filters 20a are spaced apart from each other by a predetermined distance d. As a result, since the adjacent filters 20a are arranged apart from each other, color mixing does not occur at the boundaries between the adjacent filters 20a. Therefore, quantitative expression of RGB in a captured image can be performed more reliably.

[About color filter configuration and calibration]
When estimating the direction of the surface of the article 200 described above, it is necessary that the captured image has a color balance. In other words, it is necessary that the RGB intensities in the captured image be calibrated. Generally, a white article 200 is imaged for each lighting environment, and the RGB intensities are adjusted.

Further, when the material of the article 200 is different, the wavelength dependence of the reflectance is also different. FIG. 4 shows an example of the wavelength dependence of the reflectance of metals. For example, silver (Ag) has a constant reflectance of about 1 in the visible light region (400 nm to 700 nm). On the other hand, as the wavelength ranges from 650 nm to 500 nm, the reflectance of gold (Au) decreases from about 1 to about 0.4. Further, in the wavelength range of 500 nm or less, the reflectance is approximately constant at about 0.4. As the wavelength ranges from 650 nm to 400 nm, the reflectance of copper (Cu) decreases from about 1 to about 0.5. In view of this, RGB intensity calibration is required even when the material of the article 200 to be inspected is different.

In contrast, in the inspection system 100 shown in FIG. 1A, the color filter 20 includes an R filter, a G filter, and a B filter. Therefore, it is sufficient to configure the RGB intensity according to the components (gold, copper, etc.) of the article 200 to be inspected, and the number of RGB calibrations (three times in this case) can be suppressed.

For example, let the RGB components of the test light transmitted through the R filter be (Rr, Gr, Br), let the RGB components of the test light transmitted through the G filter be (Rg, Gg, Bg), and let the RGB components of the test light transmitted through the B filter be (Rr, Gr, Br). Let the RGB components be (Rb, Gb, Bb). At this time, the RGB components (R, G, B) of the test light before correction and the RGB components (R', G', B') of the test light after correction are related to the following equation (2). Become.

Note that A satisfies the following formula (3).

As described above, by calibrating the RGB intensities in the captured image, a more accurate image of the article 200 can be obtained.

[About inspection results of goods]
Next, the inspection results of article 200 will be explained.

FIG. 5 is a schematic plan view of a color filter used during article inspection. FIG. 6 is a cross-sectional view of the article used during the article inspection. FIG. 7 is an image captured by a camera during an article inspection. In this inspection result, by using the color filter 21 of FIG. 5 and placing the article 200a of FIG. 6 on the support stand 70, the camera 80 shown in each figure of FIG. Images G1 to G12 were obtained.

As shown in FIG. 5, the color filter 20 has a blue area (B area) at the upper left of the drawing, a red area (R area) at the upper right of the drawing, and a green area (R area) at the bottom of the drawing. have. In FIG. 5, the color filter 21 is shown divided into an R area, a G area, and a B area, but in reality, the in-plane color distribution changes continuously or stepwise. ing. That is, the plurality of filters 20a are arranged in the color filter 20 so that the color distribution gradually changes when divided into regions within a predetermined range.

As shown in FIG. 6, the article 200a has a region S4 parallel to the X direction and the Y direction between the region S2 and the region S3. Therefore, in the image captured by the camera 80, the area corresponding to the area S4 becomes white.

Each figure in FIG. 7 shows captured images when the inclinations of regions S2 and S3 of the article 200a are changed. More specifically, captured images in which the inclinations of regions S2 and S3 of the article 200a with respect to the X direction are 1° to 12° are shown as captured images G1 to G12, respectively. As shown in each figure in FIG. 7, as the inclination of the regions S2 and S3 with respect to the X direction increases, the ratio of RGB reflected by the regions S2 and S3 to the half mirror 60 changes. According to this change in the ratio of RGB, it is possible to estimate the inclination of the surface of the article to be inspected.

Note that in the above embodiment, the camera 80 may be a monochrome camera, and the color filter 20 may include a monochrome filter 20a. Even in this case, it is possible to estimate the inclination of the surface of the article 200 by comparing the luminance values of the reference image and the captured image. That is, the same effects as the above embodiment can be obtained.

In addition, the surface of the article 200 can be inspected by correcting the camera image of the article 200 captured by the camera 80 using the inclination of the surface of the article 200 estimated by this inspection system. For example, when inspecting the surface of the article 200 (such as a coin), if there is a foreign object under the article 200 and the article 200 is tilted as a whole, the tilt of the article 200 is corrected and the article 200 is tilted. It becomes possible to create a camera image of the item 200 in a state where it is not covered and inspect the surface of the item 200.

(Embodiment 2)
FIG. 8 is a schematic plan view of a color filter according to the second embodiment. FIG. 9 is a diagram showing the color distribution within the light exit plane of the color filter. FIG. 10 is a diagram showing the rotation angle dependence of light brightness at the outermost periphery of the color filter. For convenience of explanation, in FIGS. 8 to 10 and the subsequent drawings, the same reference numerals are given to the same parts as in the first embodiment, and detailed explanations are omitted.

In the first embodiment, an example was shown in which the plurality of filters 20a in the color filter 20 are arranged so that the color distribution gradually changes when the filters 20a are divided into regions within a predetermined range. Furthermore, it was shown that the inclination of the surface of the article 200 can be estimated by comparing the brightness values of a captured image of the article 200 captured by irradiating the inspection light transmitted through the color filter 20 with a reference image acquired in advance. Ta.

In this embodiment, a specific color distribution imparting method for the already calibrated color filter 20 will be described.

First, the arrangement of each color filter in the color filter 20 will be explained.

As shown in FIG. 8, the color filter 20 includes a red filter 20R, a green filter 20G, and a blue filter 20B, which are each arranged periodically on a transparent plate 22. In this case, the transparent plate 22 is made of a glass substrate and is provided with a drive transistor (not shown). The filters are arranged at intervals d. This is similar to that shown in the first embodiment.

The red filter 20R transmits red light of the incident white light, the green filter 20G transmits the green light of the incident white light, and the blue filter 20B transmits the blue light of the incident white light. . Note that the arrangement shown in FIG. 8 is just an example, and the arrangement is not particularly limited thereto. Further, the wavelength ranges of red light, green light, and blue light are the same as shown in the first embodiment.

In this embodiment, as shown at 98, the color distribution on the light exit surface of the color filter 20 is changed continuously or stepwise in a clockwise direction around the center O of the color filter 20. Further, a color distribution is provided such that the saturation of the test light decreases continuously or stepwise from the outer periphery toward the center O. In other words, a color distribution is provided such that the saturation of the test light changes continuously or stepwise with respect to the distance r from the center O.

On the other hand, the inspection lights emitted from different positions on the light emitting surface satisfy the relationship shown in equation (4).

I _R +αI _G +βI _B =S...(4)
Here, I _R , I _G , and I _B are the brightness of red light, green light, and blue light, respectively. Further, α and β are each coefficients larger than zero, and in the example shown in this embodiment, α=β=1. Further, S is a constant.

In other words, the inspection light is set so that the sum of the brightness of the red light, green light, and blue light is constant regardless of the position within the light exit plane.

In addition, the inspection light is set so that when the angle θ rotated clockwise from a predetermined position around the center O changes, the brightness ratio of each of the red light, green light, and blue light changes. There is.

For example, as shown in FIGS. 9 and 10, the brightness of the red light, green light, and blue light at the outermost periphery of the color filter 20 changes depending on the angle θ. At the same time, the brightness of each of the red light, green light, and blue light also changes depending on the distance r. As the distance r becomes smaller, that is, closer to the center O, the saturation of the test light decreases. For example, when θ is 120°, the brightness I _G of green light is maximum (=S), while the brightness I _R and I _B of red light and blue light are respectively zero. However, as the distance r becomes smaller and in a region closer to the center O, the brightness of the red light and the blue light each increase from zero.

These matters are described as the relationships shown in equations (5) to (11) below.

First, the distance r is normalized. Further, the normalized amplitude m of the inspection light at the distance r is set as shown in equation (5).

m=min(R,G,B)...(5)
Here, R, G, and B are the standardized brightness I _R , I _G , and I _B of red light, green light, and blue light, respectively.

At this time, the color distribution of the color filter 20 is set so that the relationship shown in equation (6) is established between the distance r and the amplitude m.

r=VR/(VR+m)=(S-3m)/(S-2m)...(6)
Here, the variation range VR satisfies the relationship shown in equation (7).

VR=S-3m...(7)
Further, S, r, and m satisfy the relationships shown in equations (8) to (10), respectively.

S≧3m...(8)
0≦r≦1...(9)
0≦m≦1...(10)
Furthermore, from equations (6) and (7), the relationship shown in equation (11) holds true regarding the amplitude m.

m=S×((1-r)/(3-2r))...(11)
In the example shown in FIG. 10, in other words, at any position on the light exit surface of the inspection light, the ratio of the brightness of the red light, the brightness of the green light, and the brightness of the blue light varies depending on the distance r and the angle θ. It is determined. However, as described above, even if the angle θ or the distance r changes, the sum of the brightness of each of the red light, green light, and blue light remains constant.

Note that when providing a color distribution to the inspection light as shown in FIGS. 9 and 10, the number of red filters 20R, green filters 20G, and blue filters 20B arranged at each position of the color filter 20 is changed. Alternatively, at each position, the transmittance of each of the red filter 20R, green filter 20G, and blue filter 20B is changed.

Note that when a liquid crystal color filter is used as the color filter 20, the above-mentioned arrangement ratio and transmittance ratio can be dynamically changed.

Furthermore, by imparting a color distribution to the inspection light as shown in FIGS. 9 and 10, the inclination of the surface of the article 200 can be quantitatively evaluated. This inclination is expressed as an inclination angle based on an image of the article 200 captured by the camera 80 with reference to the case where the surface of the article 200 is a flat surface perpendicular to the optical axis of the inspection light. Note that the angle of inclination of the surface of the article 200 (hereinafter sometimes simply referred to as the angle of inclination) is calculated by a control device (not shown). Further, by comparing the brightness difference between the reference image and the captured image, the inclination of the surface of the article 200 can be quantitatively evaluated in the same manner in the first embodiment.

As explained above, in this embodiment, the color filter 20 is given a color distribution as shown in FIGS. 9 and 10.

In this case, when the respective brightnesses of the red light, green light, and blue light are I _R , I _G , and I _B at any position within the light irradiation surface of the inspection light,
I _R +αI _G +βI _B =S...(4)
The relationship shown in is satisfied. Here, α and β are each coefficients larger than zero, and S is a constant.

By doing so, the color balance of the image captured by the camera 80 can be appropriately adjusted even when the surface of the article 200 has multiple directions or when the article 200 is made of different materials. Further, it becomes possible to quantitatively and accurately calculate the direction of the surface of the article 200, specifically, the angle of inclination of the surface.

When the color filter 20 is circular in plan view, the saturation of the test light preferably changes such that the saturation decreases from the outer periphery of the color filter 20 toward the center O.

Furthermore, the brightness I _R , I _G , and I _B of the red light, green light, and blue light preferably vary depending on the rotation angle θ along the outer periphery of the surface light source 10 .

Furthermore, the sum of the luminances of the red light, green light, and blue light is S, the standardized distance from the center O of the color filter 20 is r, and the standardized amplitude of the inspection light at the distance r is m. shall be. At this time,
m=S×((1-r)/(3-2r))...(11)
It is preferable that the following relationship is satisfied. Here, S≧3m, 0≦r≦1, and 0≦m≦1.

By doing so, it is possible to give the inspection light a color distribution that satisfies the relationship shown in equation (4) at any position within the light irradiation surface of the inspection light.

The ratio of the number of arranged red filters 20R, green filters 20G and blue filters 20B (arranged number ratio) or transmittance ratio is determined so as to satisfy the relationships shown in equations (4) and (11), respectively. preferable.

By doing so, the color balance of the image captured by the camera 80 can be appropriately adjusted even when the surface of the article 200 has multiple directions or when the article 200 is made of different materials. Furthermore, the color distribution of the test light, specifically the RGB brightness ratio, can be continuously changed within the light exit plane of the test light.

Furthermore, it is possible to quantitatively and accurately calculate the direction of the surface of the article 200, specifically, the inclination angle of the surface.

<Modified example>
FIG. 11 is a schematic diagram illustrating the color distribution of inspection light according to a modification. FIG. 12A is a schematic diagram showing the color distribution of blue light in the inspection light. FIG. 12B is a schematic diagram showing the color distribution of red light in the inspection light.

In imparting a color distribution to the inspection light so as to satisfy the relationship shown in equation (4), the shape of the color filter 20 is not limited to that shown in FIGS. 5 and 9, but may be, for example, as shown in FIG. Additionally, it may be rectangular in plan view.

In this case as well, a color distribution is given to the inspection light so as to satisfy the relationship shown in equation (4). However, in this modification, α=2, β=1, and S=2 are set.

I _R +2I _G +I _B =2...(4A)
Further, along one side of the color filter 20, the brightness _IB of the blue light changes continuously or stepwise, and along the other side orthogonal to the one side, the brightness _IR of the red light changes. A color distribution is given to the inspection light so that it changes continuously or stepwise.

In the example shown in FIG. 11, the brightness _IB of blue light increases continuously or stepwise along the Y direction, and the brightness _IR of red light increases continuously or stepwise along the X direction. ing. Specifically, as shown in FIGS. 12A and 12B, the brightness distributions of blue light and red light are respectively set.

Note that the brightness _IG of the green light is changed within the plane of the surface light source 10 so as to satisfy equation (2A). For example, at the center of the color filter 20, I _R :I _G :I _B =0.5:0.5:0.5, and at the tip of the virtual line shown by the broken line (arrow portion), I _R :I _G : The brightness I _G of the green light is set so that I _B =1:0:1, and I _R :I _G :I _B =0:1:0 at the proximal end.

That is, in the color filter 20 shown in this modification, one color is selected as the first color from red light, green light, and blue light. Among the respective brightnesses of the red light, green light, and blue light, the brightness of the first color changes continuously or stepwise along one side of the color filter 20.

Among the red light, green light, and blue light, the brightness of the second color excluding the first color changes continuously or stepwise along the other side that intersects one side.

According to this modification, even if the color filter 20 has a rectangular shape unlike the second embodiment, the same effects as the configuration shown in the second embodiment can be achieved. That is, even when the surface of the article 200 has a plurality of directions or the article 200 is made of different materials, the color balance of the image captured by the camera 80 can be appropriately adjusted. Further, it becomes possible to quantitatively and accurately calculate the direction of the surface of the article 200, specifically, the angle of inclination of the surface.

FIG. 13 is a schematic configuration diagram of an inspection system according to the third embodiment. FIG. 14 is a schematic diagram of an image of an article when the article is not correctly installed on the support stand. The inspection system 100 shown in FIG. 13 differs from the inspection system 100 shown in FIG. 1A only in that the article 200 is not correctly installed on the support stand 70.

If the article 200 is not correctly installed on the support stand 70, an offset is added to the calculated inclination angle, and the correct value may not be obtained. Further, there may be cases where it is desired to remove the inclination of the article 200 as an offset.

In such a case, the inclination angle can be calculated accurately by removing the offset using the procedure shown below.

As shown in FIG. 13, when the article 200 is installed at an angle with respect to the support stand 70, even if the surface of the article 200 is flat or has a certain degree of flatness with some unevenness, the image of the article 200 is , is colored in a color corresponding to the inclination with respect to the support stand 70. When an image of the article 200 is acquired in the installed state shown in FIG. 13, the image of the article 200 shown in FIG. 14 is bluish.

Here, a point on the surface of the article 200 that will serve as a reference (hereinafter referred to as a reference surface) is set. In the example shown in FIG. 14, the plane including the cross mark 210 is set as the reference plane. This reference surface is preferably a location that is known in advance to be flat. Furthermore, in order to extract the reference plane from the image of the article 200, it is preferable that the reference plane includes a characteristic shape, as shown in FIG. It goes without saying that the characteristic shape is not limited to the cross mark 210.

Next, an offset is calculated based on the image of the reference plane, specifically, the color of the image, the intensity of its brightness, etc.

Further, the previously determined offset is uniformly subtracted from the inclination angle of each part of the article 200, which is calculated based on the image of the article 200.

By doing so, the offset superimposed on the calculated inclination angle can be appropriately removed, and the inclination angle of the surface of the article 200 can be accurately determined. Note that the offset described using FIGS. 13 and 14 corresponds to the inclination of the article 200 with respect to the support base 70. Including the inclination of the article 200 itself, the offset in this specification is defined as the inclination of the article 200 with respect to the inspection system 100. Note that when the surface of the article 200 has a plurality of directions, the inclination of the article 200 itself includes the inclination of any arbitrary location within the article 200, the average of the inclinations of multiple locations within the article 200, etc. .

As explained above, the method for correcting the inclination angle of the surface of the article 200 according to the present embodiment includes the following first to fourth steps.

In the first step, the inspection light is irradiated onto the article 200, and an image of the article 200 is acquired by the camera 80.

In the second step, a reference plane is set on the surface of the article 200.

In the third step, the offset is calculated based on the image of the reference plane. This offset corresponds to the tilt of article 200 with respect to inspection system 100.

In the fourth step, the above-mentioned inclination angle is corrected by subtracting the offset calculated in the third step from the inclination angle at each part of the surface of the article 200 obtained in the first step.

In the actual appearance inspection of the article 200, the article 200 may not be correctly installed on the support stand 70 due to reasons such as the adjustment of the transportation system being slightly off, or vibrations being applied to the article 200 during transportation and installation. be.

According to the present embodiment, even if the article 200 is not correctly installed on the support stand 70, the influence of this is removed, the direction of the surface of the article 200 is correctly estimated, and the inclination angle is accurately determined. It can be calculated. Furthermore, the surface of the article 200 can be easily observed by removing the inclination that the article 200 itself has. For example, when the surface of the article 200 has multiple directions, the appearance of the article 200 can be easily inspected by removing inclinations at desired locations within the article 200. Alternatively, as the inclination of the article 200 itself, the average value of the inclinations at multiple locations within the article 200 may be selected, and the average value may be subtracted from the inclination angle at each part of the surface of the article 200.

Note that the method for correcting the inclination angle of the surface of the article 200 is not particularly limited to this. For example, the inspection light is made monochromatic and irradiated onto the article 200, and an image thereof is acquired. Furthermore, the color distribution of the inspection light is changed to, for example, the pattern shown in FIG. 9, and the article 200 is irradiated to obtain an image thereof. Color correction is performed on the latter image using the former image obtained by irradiating monochromatic inspection light. The inclination angle of the surface of the article 200 is calculated based on the image after color correction. In the following description, the test light that is monochromatic light may be referred to as a first test light, and the test light provided with a color distribution may be referred to as a second test light. The second inspection light includes at least light in the same wavelength band as the monochromatic light and light in a different wavelength band from the monochromatic light. By doing so, the inclination angle of the surface of the article 200 can be accurately calculated.

Note that in color-calibrating the inspection light, it is preferable to use images obtained by irradiating three types of monochromatic inspection light: red light, green light, and blue light.

(Other embodiments)
Even when the color filter 20 shown in the second embodiment is used, estimating the direction of the surface of the article 200 and calculating the inclination angle based on the comparison between the reference image and the captured image is different from the one shown in the first embodiment. It is similar to Therefore, the inspection system 100 shown in this specification determines the angle and/or direction of the surface of the article 200 with respect to the image of the article 200 captured by the camera 80, depending on the components of the inspection light that has passed through the plurality of filters. Calculate quantitatively. In other words, the inspection system 100 quantitatively determines at least one of the angle and direction of the surface of the article 200 with respect to the image of the article 200 captured by the camera 80 according to the components of the inspection light that has passed through the plurality of filters. calculate.

The inspection system of the present disclosure is capable of estimating the direction of the surface of the article and adjusting the color balance of the captured image, so it is useful for use in inspecting the appearance of the article.

10 Surface light sources 20, 21 Color filter 20a Filter 20R Red filter 20G Green filter 20B Blue filter 30 Lens 40 Housing 50 Illumination device 60 Half mirror 70 Support stand 80 Camera (imaging device)
100 Inspection system 200, 200a Article 210 Cross mark

Claims

An inspection system comprising at least an illumination device that irradiates an article with inspection light, and an imaging device that images the article irradiated with the inspection light,
The lighting device includes:
a light source that emits the inspection light;
a color filter that is disposed between the light source and the article in the optical path of the inspection light and includes a plurality of filters that transmit light of mutually different wavelength widths;
The inspection system is characterized in that the inclination of the surface of the article is detected based on an image of the article captured by the imaging device and a reference image.
The inspection system according to claim 1,
The reference image is a captured image of the article that has been determined in advance to have no abnormality,
The inspection system is characterized in that the inspection system detects the inclination of the surface of the article or the presence or absence of an abnormality on the surface of the article based on the difference between the image of the article newly captured by the imaging device and the reference image. inspection system.
The inspection system according to claim 1,
The inspection system is characterized in that the plurality of filters are arranged on the color filter such that adjacent filters are spaced apart from each other by a predetermined distance.
The inspection system according to claim 1,
The inspection system is characterized in that the plurality of filters are arranged in an array on the color filter, and when divided into regions within a predetermined range, the color distribution gradually changes.
The inspection system according to claim 1,
An inspection system characterized in that an image of the article captured by the imaging device is corrected according to components of the inspection light that has passed through the plurality of filters.
The inspection system according to claim 3,
The color filter includes a red filter that transmits red light, a green filter that transmits green light, and a blue filter that transmits blue light that are arranged in a plane,
When the respective brightnesses of the red light, the green light, and the blue light are I R , I G , and I B at any position within the light irradiation surface of the inspection light,
I R +αI G +βI B =S...(4)
An inspection system characterized by satisfying the relationship shown in .
Here, α and β are each coefficients larger than zero, and S is a constant.
The inspection system according to claim 6,
When the color filter is circular in plan view,
The chroma of the test light changes such that the chroma decreases from the outer periphery toward the center of the color filter,
An inspection system characterized in that the brightness of each of the red light, the green light, and the blue light changes according to a rotation angle along the outer periphery of the color filter.
The inspection system according to claim 7,
Let S be the sum of the respective luminances of the red light, the green light, and the blue light,
Let r be the normalized distance from the center of the color filter,
When the normalized amplitude of the inspection light at the distance r is defined as m,
m=S×((1-r)/(3-2r))...(11)
An inspection system characterized by satisfying the relationship shown in .
Here, S≧3m, 0≦r≦1, and 0≦m≦1.
The inspection system according to claim 6,
When the color filter is rectangular in plan view,
Among the respective brightnesses of the red light, the green light, and the blue light, the brightness of the first color changes continuously or stepwise along one side of the color filter,
Among the red light, the green light, and the blue light, the brightness of a second color other than the first color changes continuously or stepwise along another side that intersects the one side. An inspection system characterized by:
The inspection system according to claim 1,
At least one of an angle and a direction of a surface of the article is quantitatively calculated with respect to an image of the article captured by the imaging device, according to components of the inspection light transmitted through the plurality of filters. inspection system.
A method for correcting the tilt angle of the surface of an article using the inspection system according to claim 1, comprising:
a first step of irradiating the article with the inspection light and acquiring an image of the article with the imaging device;
a second step of setting a reference plane on the surface of the article;
a third step of calculating an offset based on the image of the reference plane;
At least a fourth step of correcting the inclination angle by subtracting the offset calculated in the third step from the inclination angle at each part of the surface of the article obtained in the first step,
A method for correcting an inclination angle of a surface of an article, wherein the offset corresponds to an inclination of the article with respect to the inspection system.