JP2004361085A - Visual examination device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a visual examination device capable of discriminating the surface unevenness of translucent matter or the air bubbles in the translucent matter from another flaw to detect the same. <P>SOLUTION: The visual examination device is equipped with a visualization means for visualizing the surface unevenness of the translucent matter or the air bubbles in the translucent matter as pair of light and darkness parts, an imaging means for imaging the light and darkness parts visualized by the visualization means, a light and darkness extraction means for extracting the light part and darkness part of the image acquired by the imaging means and a judge means for judging the light and darkness parts as unevenness or air bubbles when the distance between the light part and the darkness part extracted by the light and darkness extraction means is smaller than a predetermined value. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a visual inspection device that detects irregularities on the surface of an object and bubbles inside a transparent object.
[0002]
[Prior art]
As a means for visualizing irregularities on the surface of an object or bubbles inside a transparent object, there is known a method of irradiating an inspected object with light to visualize the irregularities and bubbles as a pair of bright and dark portions (for example, see Patent Document 1). ).
[0003]
Further, as a means for detecting the visualized irregularities and bubbles, there is known a method of comparing an inspected object with a non-defective product and detecting a portion having a difference in brightness as irregularities or bubbles (for example, see Patent Document 2). ).
[0004]
[Patent Document 1]
JP-A-2002-310939.
[0005]
[Patent Document 2]
JP-A-9-193515.
[0006]
[Problems to be solved by the invention]
However, since the irregularities and bubbles visualized by the above-described method are constituted by a pair of bright portions and dark portions, one irregularity or bubbles may be detected as two defects, a bright defect and a dark defect.
[0007]
In addition, irregularities on the surface of the object and bubbles inside the transparent object have low contrast, and it is necessary to increase the detection sensitivity in order to avoid oversight. However, when the detection sensitivity is increased, a normal portion is also detected as a defect, that is, so-called overdetection is likely to occur.
[0008]
The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a visual inspection device capable of detecting irregularities on the surface of an object and bubbles inside a translucent object separately from other defects. It is in.
[0009]
[Means for Solving the Problems]
In order to solve the above problems and achieve the objects, a visual inspection device of the present invention is configured as follows.
[0010]
(1) Visualization means for visualizing irregularities on the surface of an object or bubbles inside a transparent object as a pair of light and dark parts, imaging means for imaging the light and dark parts visualized by the visualization means, and an image taken by the imaging means Light and dark extraction means for extracting the light and dark parts by image processing, and when the distance between the light and dark parts extracted by the extraction means is smaller than a predetermined value, the light and dark parts are determined to be the irregularities or bubbles. And a judging means.
[0011]
(2) In the appearance inspection apparatus described in (1), the light / dark extraction unit extracts a region exceeding a first gradation in an image captured by the imaging unit as a bright defect region, A region less than the second gradation lower than the first gradation is extracted as a dark defect region.
[0012]
(3) The visual inspection apparatus according to (2), wherein the bright defect area is enlarged to form an enlarged bright defect area, and the dark defect area is enlarged to form an enlarged dark defect area. Means, and an adding means for adding the enlarged bright defect area and the enlarged dark defect area formed by the enlarging means, wherein the judging means comprises adding the enlarged bright defect area and the enlarged dark defect area. Thus, when these areas are connected, it is determined that the distance between the bright defect area and the dark defect area is smaller than a predetermined value.
[0013]
(4) In the appearance inspection device described in (3), the enlargement unit enlarges the bright defect area and the dark defect area in all directions around the defect area.
[0014]
(5) In the appearance inspection apparatus described in (3), the visualization unit visualizes the irregularities and bubbles such that the direction of contrast of the bright and dark portions is constant.
[0015]
(6) In the appearance inspection apparatus described in (5), the enlargement unit enlarges the bright defect area and the dark defect area in directions approaching each other.
[0016]
(7) In the appearance inspection device described in (5), the determination unit may be configured to remove the light and dark portion from the unevenness or the unevenness when the gradation of the light and dark portion smoothly changes in the contrast direction. It is characterized in that it is determined to be a bubble.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, a first embodiment of the present invention will be described with reference to FIGS.
[0018]
FIG. 1 is a schematic diagram showing a configuration of a visual inspection device according to a first embodiment of the present invention.
[0019]
The appearance inspection apparatus shown in FIG. On the upper surface of the mounting stage 1, the inspection object A is mounted. Immediately above the inspection object A mounted on the mounting stage 1, a camera 2 (imaging means) is fixed with its lens (negative illustration) facing the inspection object A side. The camera 2 captures an image of the inspection object A on the mounting stage 1.
[0020]
An illumination device 3 (visualization means) is provided at a position slightly away from the camera 2 above the inspection object A mounted on the upper surface of the mounting stage 1. The illumination device 3 irradiates light to the inspected object A with the optical axis inclined by an angle θ.
[0021]
The visualization means may be, for example, a ring-shaped illumination horizontally arranged above the inspected object A, other than the illumination device 3 which irradiates light from one direction as shown in FIG. Further, instead of the illumination device 3, a differential interference microscope or the like may be used.
[0022]
A computer 4 is connected to the camera 2. The computer 4 has an image capturing device 5, a storage device 6, and a computing device 7 (light / dark extraction means, enlargement means, addition means, judgment means, certainty means), takes in an image taken by the camera 2, And process.
[0023]
Next, an operation when the visual inspection device having the above configuration is used will be described.
[0024]
FIG. 2 is an image M to be inspected obtained by capturing the object A to be inspected used in the embodiment by the image capturing device 5.
[0025]
The inspected object A is formed of a transparent material, and has bubbles B mixed therein. When the inspection object A is placed on the upper surface of the mounting stage 1, the angle θ of the optical axis of the illumination device 3 is adjusted, and light is emitted toward the inspection object A. As a result, the bubbles B mixed into the inspection object A are visualized as a pair of bright and dark portions.
[0026]
When the bubbles B are visualized as light and dark portions, the object 2 to be inspected is imaged by the camera 2. The image captured by the camera 1 is captured by the image capturing device 5 as the inspection target image M as shown in FIG. Then, the inspection target image M is taken out from the storage device 6 as needed, and the calculation device 7 detects the bubbles B.
[0027]
Hereinafter, the process of detecting the bubble B will be described.
[0028]
3A and 3B show a process of detecting a defect in the inspection object A using the inspection object image M according to the embodiment, wherein FIG. 3A is a process diagram when the defect is a bubble B, and FIG. FIG. 4 is a process chart in the case where the defect is another defect, and FIG. 4 is a grid diagram showing a binarized image of the inspection target image M according to the embodiment, where (a) shows a bright defect display image (dark defect display). (B) are enlarged bright defect display images (enlarged dark defect display images).
[0029]
As shown in FIG. 3A, the inspection target image M captured by the image capturing device 5 is binarized by the arithmetic unit 7 using the first gradation R and the second gradation L as thresholds. (Light / dark portion extraction processing). Note that the second gradation L is set lower than the first gradation R.
[0030]
When the inspection target image M is binarized using the first gradation R as a threshold, pixels exceeding the first gradation R are represented by 1 (displayed in white in FIG. 3A), and the other pixels are represented by white. The area indicated by 0 (displayed in black in FIG. 3A) and the area indicated by 1 are defined as bright defect areas 51. The image obtained by this is defined as a bright defect display image.
[0031]
On the other hand, when the image to be inspected M is binarized using the second gradation L as a threshold, pixels that do not satisfy the second gradation L are 1 (displayed in white in FIG. 3A), Are indicated as 0 (displayed in black in FIG. 3A), and the area indicated by 1 is defined as a dark defect area 52. The image obtained by this is defined as a dark defect display image.
[0032]
When the bright defect display image and the dark defect display image are obtained, the arithmetic unit 6 enlarges the bright defect region 51 and the dark defect region 52 in each image by one pixel in all directions around the defect (defect enlargement processing 1). That is, as shown in FIG. 4, the display of the minimum pixels surrounding the bright defect area 51 and the dark defect area 52 is changed from 0 to 1.
[0033]
Then, the area displayed as 1 in the bright defect display image due to the enlargement of the bright defect area 51 is referred to as an enlarged bright defect area 53, and the area indicated as 1 in the dark defect display image is enlarged due to the enlargement of the dark defect area 52. The dark defect area 54 is set. The images obtained in this way are referred to as an enlarged bright defect display image and an enlarged dark defect display image, respectively.
[0034]
When the enlarged bright defect display image and the enlarged dark defect display image are formed, the logical sum of the truth value of the pixels constituting the enlarged bright defect display image and the truth value of the pixels constituting the enlarged dark defect display image is displayed as 1. A connected dark / dark defect image showing the area to be connected as a connected defective area 55 is constructed. That is, the connected image of light and dark defects includes an area included in at least one of the enlarged light defect area 53 and the enlarged dark defect area 54.
[0035]
At this time, it is determined whether or not the enlarged bright defect area 53 and the enlarged dark defect area 54 are connected, that is, whether or not the connected defect area 55 forms one continuous area, and forms one area. In such a case, it is determined that the light and dark portions constituting the light and dark portions are within a predetermined distance range, that is, close to each other (defect determination process).
[0036]
Generally, when the bubble B is visualized as a pair of bright and dark portions by light irradiation, the bright portion and the dark portion are extremely close to each other. Therefore, in the present embodiment, when the bright portion and the dark portion are within the predetermined range as described above, it is determined that the bright and dark portion is caused by one bubble B.
[0037]
On the other hand, if the enlarged bright defect area 53 and the enlarged dark defect area 54 are not connected, that is, if the connected defective area 55 does not become one continuous area, the above-described bright / dark area is determined as two adjacent independent defects (defects). Determination process). Such a defect may be adhesion of dust or the like.
[0038]
Also, as shown in FIG. 3A, even when only a bright portion or a dark portion is visualized alone in the inspection target image M, it is determined that the bright portion or the dark portion is caused by adhesion of dust or the like (defects). Determination process).
[0039]
According to the appearance inspection apparatus having the above-described configuration, the bubbles B visualized as a pair of light and dark portions are extracted as the light defect region 51 and the dark defect region 52, and these regions are enlarged by one pixel in all directions in the surroundings. A defect area 53 and an enlarged dark defect area 54 are provided.
[0040]
Then, it is determined whether or not the distance between the bright defect area 51 and the dark defect area 52 is within a predetermined range by calculating the logical sum of the enlarged bright defect area 53 and the enlarged dark defect area 54. .
[0041]
Only when the pair of the light defect area 51 and the dark defect area 52 constituting the light and dark area is within a predetermined range, the light and dark area is detected as one bubble B.
[0042]
Therefore, the bright and dark portions visualized in the inspection target image M by the bubbles B can be detected separately from the parts visualized in the inspection target image M due to the attachment of dust or the like.
[0043]
Furthermore, even if a pair of bright portions and dark portions are visualized as a pair of bright portions and dark portions in the image M to be inspected due to the attachment of two dusts or the like, it can be prevented that these portions are erroneously detected as bubbles B. .
[0044]
In the present embodiment, the detection target is the bubble B mixed into the inspected object A, but the present invention can also be applied to irregularities formed on the surface of the object. In this case, the inspection object A does not need to be transparent.
[0045]
Next, a second embodiment of the present invention will be described with reference to FIGS.
[0046]
Note that description of the same configuration, steps, and operation as in the first embodiment will be omitted.
[0047]
FIG. 5 is a process diagram showing a process of detecting a bubble B in an inspection object A using an inspection object image M according to the second embodiment of the present invention, and FIG. 6 is an inspection process according to the second embodiment. FIGS. 7A and 7B are grid diagrams showing the target image M in a binarized manner, wherein FIG. 7A shows a bright defect display image, FIG. 7B shows an enlarged bright defect display image, and FIG. It is a grid figure shown as a value, (a) is a dark defect display image, (b) is an enlarged dark defect display image.
[0048]
In the present embodiment, the lighting device 3 used in the first embodiment is used. Since the illuminating device 3 irradiates light to the inspection object A from one direction as described above, the direction of the light and dark contrast formed on the inspection object A is over the entire surface of the inspection object A. It becomes almost uniform.
[0049]
Therefore, in the present embodiment, when the bright defect area 51 is enlarged, the bright defect area 51 is enlarged one pixel at a time toward the dark defect area 52 paired with the bright defect area 51 along the direction of light / dark contrast. When the dark defect area 52 is enlarged, the pixel is enlarged one pixel at a time along the direction of the contrast between light and dark toward the bright defect area 51 that is a pair with the dark defect area 52 (defect enlargement processing 2).
[0050]
That is, as shown in FIGS. 6 and 7, when the contrast between light and dark is in the direction of arrow C, the bright defect area 51 is enlarged by one pixel in the direction of arrow C, and the dark defect area 52 is expanded by one pixel in the direction opposite to arrow C. Each pixel is enlarged.
[0051]
The image obtained by enlarging the bright defect area 51 as a partially enlarged bright defect area 61 is an image showing a partially enlarged bright defect display image, and the image obtained by enlarging the dark defect area 52 is shown as a single enlarged dark defect area 62. Is a half-enlarged dark defect display image (see FIG. 5).
[0052]
As described above, by enlarging the bright defect area 51 and the dark defect area 52 in consideration of the direction of the contrast between light and dark, a portion visualized in the inspection target image M due to the attachment of dust or the like is detected as a bubble B. The false detection rate can be reduced.
[0053]
Next, a third embodiment of the present invention will be described with reference to FIGS.
[0054]
Here, the description of the same configuration, process, and operation as those of the first embodiment and the second embodiment will be omitted.
[0055]
FIG. 8 is a process diagram showing a process of detecting bubbles B in the inspection object A using the inspection image M according to the third embodiment of the present invention, and FIG. It is a graph which shows the value (differential value) differentiated in the direction of the contrast of light and dark.
[0056]
In the present embodiment, similarly to the second embodiment, light is emitted from one direction to the inspection object A using the illumination device 3. Therefore, the direction of light / dark contrast formed on the inspected object A is substantially uniform over the entire surface of the inspected object A.
[0057]
When the bright defect display image and the dark defect display image are obtained, the logical value of the truth value of the pixels constituting the bright defect display image and the truth value of the pixels constituting the dark defect display image is calculated, and the area indicated by 1 is calculated. A bright / dark defect display image shown as the bright / dark defect area 58 is configured.
[0058]
On the other hand, when the bright defect display image and the dark defect display image are obtained, the half-enlarged bright defect display image (negative illustration) and the half-enlarged dark defect display image (negative illustration) are obtained in the same manner as in the second embodiment. (Defect enlargement processing 2), and the one-side enlarged bright defect display image and the one-side enlarged dark defect display image are combined to form a combined bright / dark defect image.
[0059]
On the other hand, the gradation of the image M to be inspected taken into the image taking device 6 is differentiated in the direction of light / dark contrast by the computing device 7 to form a differential image showing the differential value as light / dark.
[0060]
In general, when the bubble B is visualized as a pair of light and dark parts by light irradiation, the gradation at the connecting part of the light part and the dark part changes smoothly. Therefore, as shown in FIG. 9, the differential value of the gradation at the connecting portion between the bright portion and the dark portion slopes smoothly. Moreover, in the present embodiment, since the direction of the contrast between light and dark is made constant and the bright part and the dark part are visualized in the order of the contrast in the direction of the contrast, the differential value at the connection part takes a negative value. .
[0061]
Next, using this differential image, a differential value negative part extraction image displaying a pixel having a negative differential value as 1 (shown in white in FIG. 8) and the other pixels as 0 (shown in black in FIG. 8) (Negative value part extraction processing). As a result, an area in which the gradation at the connecting part of the light part and the dark part smoothly changes from light to dark is indicated by 1 (white). This area is referred to as a negative value area 56.
[0062]
Then, the logical value of the truth value of the pixel forming the differential value negative portion extracted image and the truth value of the pixel forming the above-described bright / dark defect connected image is calculated, and the area indicated by 1 (white) is determined as the defect connecting portion. The image of the defect connection portion as the region 57 is formed.
[0063]
That is, the defect connection portion region 57 is a region where the light defect region 51 and the dark defect region 52 are within a predetermined range, and the gradation at the connection portion between the light portion and the dark portion changes smoothly.
[0064]
Next, the logical value of the truth values of the pixels constituting the defect connection portion image and the truth values of the pixels constituting the above-described bright / dark defect display image is calculated, and the region indicated by 1 is defined as the bright / dark defect negative value region 59. To form a connected image of a negative value portion of a light and dark defect.
[0065]
At this time, it is determined whether or not the defect connection portion area 57 and the light and dark defect area 58 are connected, that is, whether or not the light and dark defect negative value area 59 forms one continuous area.
[0066]
When the light-and-dark defect negative value part area 59 forms one area, the light part and the dark part constituting the light-dark part are close to each other, and the defect connecting part area 57 has both the light part and the dark part. , And the gradation at the connecting portion between the bright portion and the dark portion changes smoothly, so that it is determined that the bright and dark portion is caused by one bubble B.
[0067]
According to the present embodiment, the defect detected as the bubble B in the second embodiment is further inspected by using the differential value of the image M to be inspected. That is, the gradation in the portion of the bubble B is differentiated in the direction of light / dark contrast, and it is determined whether or not the differential value in the portion recognized as the bubble B in the second embodiment smoothly slopes toward negative. ing.
[0068]
In other words, even when two adjacent dusts or the like are visualized as a bright portion and a dark portion in the direction of the contrast, there is a small distance between the dusts, so that the portion where the differential value takes 0 is small. Should exist.
[0069]
However, in the present invention, since it is determined whether or not the differential value is smoothly negative, it is possible to improve the reliability that the target light and dark portion indicates the bubble B.
[0070]
It should be noted that the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the components without departing from the scope of the invention at the stage of implementation. Various inventions can be formed by appropriately combining a plurality of components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiment. Further, components of different embodiments may be appropriately combined.
[0071]
【The invention's effect】
ADVANTAGE OF THE INVENTION According to this invention, the unevenness | corrugation of an object surface and the bubble inside a translucent object can be detected separately from another defect.
[Brief description of the drawings]
FIG. 1 is a schematic diagram showing a configuration of a visual inspection device according to a first embodiment of the present invention.
FIG. 2 is a diagram showing an image to be inspected obtained by capturing an object to be inspected used in the embodiment by an image capturing device.
3A and 3B show a process of detecting a defect in an object to be inspected using an image to be inspected according to the embodiment, wherein FIG. 3A is a process diagram when the defect is a bubble, and FIG. FIG.
FIG. 4 is a grid diagram showing a binarized image of the inspection object according to the embodiment, wherein (a) is a bright defect display image (dark defect display image), and (b) is an enlarged bright defect display image. (Enlarged dark defect display image).
FIG. 5 is a process diagram showing a process of detecting a bubble in an object to be inspected using an image to be inspected according to a second embodiment of the present invention.
FIG. 6 is a grid diagram showing a binarized image of the inspection target image according to the embodiment, wherein (a) is a bright defect display image, and (b) is an enlarged bright defect display image.
FIG. 7 is a grid diagram showing a binarized image of the inspection target image according to the embodiment, wherein (a) is a dark defect display image, and (b) is an enlarged dark defect display image.
FIG. 8 is a process chart showing a process of detecting a bubble in a test object using a test image according to a third embodiment of the present invention.
FIG. 9 is a graph showing values obtained by differentiating bubbles in an image to be inspected in the direction of light-dark contrast.
[Explanation of symbols]
Reference numeral 2: camera (imaging means), 3: illumination device (visualization means), 7: arithmetic device (light / dark extraction means, judgment means, addition means, certainty means, enlargement means), 51: light defect area, 52: dark defect area 55, connected defect area, 61, enlarged bright defect area, 62, enlarged dark defect area, A: inspected object (object, transparent object), B: bubble, C: direction of contrast, R: first gradation , L: second gradation, M: inspection target image.

Claims (7)

Visualization means for visualizing irregularities on the object surface and bubbles inside the transparent object as a pair of light and dark portions,
Imaging means for imaging the light and dark areas visualized by the visualization means;
A light-dark extraction means for extracting a bright part and a dark part of an image taken by the imaging means by image processing;
Judging means for judging the light and dark portions as the irregularities or bubbles when the distance between the light and dark portions extracted by the light and dark extracting means is smaller than a predetermined value;
A visual inspection device comprising: The light / dark extraction means extracts a region exceeding a first gradation in an image picked up by the image pickup means as a light defect region, and extracts an area less than a second gradation lower than the first gradation. 2. The visual inspection apparatus according to claim 1, wherein the visual inspection apparatus extracts the dark defect area. Enlarging means for enlarging the bright defect area to form an enlarged bright defect area, and enlarging the dark defect area to form an enlarged dark defect area;
Adding means for adding the enlarged bright defect area and the enlarged dark defect area formed by the enlarging means;
Further comprising
The determining means determines that the distance between the bright defect area and the dark defect area is smaller than a predetermined value when the enlarged bright defect area and the enlarged dark defect area are added and these areas are connected. The appearance inspection apparatus according to claim 2, wherein: 4. The visual inspection apparatus according to claim 3, wherein said enlargement means enlarges each of the bright defect area and the dark defect area in all directions around the bright defect area and the dark defect area. 4. The visual inspection apparatus according to claim 3, wherein the visualizing means visualizes the irregularities and bubbles so that the contrast direction of the light and dark portions is constant. 6. The visual inspection apparatus according to claim 5, wherein the enlargement unit enlarges the bright defect area and the dark defect area in directions approaching each other. 6. An appearance inspection apparatus according to claim 5, wherein said determining means determines said bright and dark portions as said irregularities or bubbles when the gradation of said bright and dark portions smoothly changes in the direction of said contrast. .

Images