JP3409695B2 - Surface defect inspection method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for inspecting a coated surface for defects.

[0002]

2. Description of the Related Art Conventional surface inspection defect methods include, for example,
As disclosed in Japanese Utility Model Laid-Open No. 62-28155, slit light is projected onto the inspection surface of the surface of the object to be inspected, the reflected light is projected onto the screen, and the image on the screen is passed through a condenser lens. A method of detecting with a line sensor is known. When there is a defect on the surface of the object to be inspected, a cut image is taken on the screen at a position corresponding to the defect.

[0003]

However, according to the conventional surface defect inspection method, even a defect due to the surface roughness of the surface to be inspected such as orange peel may be judged as a defect. There was a problem. The purpose of the present invention is to
It is an object of the present invention to provide a surface defect inspection method which does not judge defects due to the surface roughness of the surface to be inspected.

[0004]

The present invention which achieves the above object is as follows. (1) Each dark portion have a plurality of Koritsu the dark portion and irradiation step Projects a light source <br/> Ru dark pattern to have a periphery to the inspected surface, the images of the specimen surface bright and dark pattern is projected An image capturing step of capturing into an image processing apparatus, a binarizing step of converting the image captured in the image capturing step into a bright and dark binary image, and an image binarized by the binarizing step. The area / center of gravity extraction step for extracting the area and center of gravity of each dark area, and the area extracted by the area / center of gravity extraction step is separated from each dark area having a predetermined area or more by a predetermined distance or more. A surface defect inspection method, which comprises: a determination step of determining a dark region at a different position as a defect. (2) Each dark portion have a plurality of Koritsu the dark portion and irradiation step Projects a light source <br/> Ru dark pattern to have a periphery to the inspected surface, the images of the specimen surface bright and dark pattern is projected An image capturing step of capturing into an image processing apparatus, a binarizing step of converting the image captured in the image capturing step into a bright and dark binary image, and an image binarized by the binarizing step. Of the area extracted by the area extracting step and the area extracting step of extracting the area of each dark area after the expanding step, and having an area smaller than a predetermined area And a step of determining a dark region as a defect, the surface defect inspection method. (3) Each dark portion have a plurality of Koritsu the dark portion and irradiation step Projects a light source <br/> Ru dark pattern to have a periphery to the inspected surface, the images of the specimen surface bright and dark pattern is projected An image capturing step of capturing into an image processing apparatus, a binarizing step of converting the image captured in the image capturing step into a bright and dark binary image, and an image binarized by the binarizing step. A shrinking step of shrinking the dark area of the image, an expanding step of expanding the dark area of the image after the shrinking step, a first inversion step of inverting the light and dark of the image after the expanding step, and the shrinking A first logical product extracting step for extracting a logical product of the image after the step and the image after the first inverting step, and a step for inverting the light and dark of the image after the first logical product extracting step Extract the logical product of the second inversion process, the image after the second inversion process, and the image after the binarization process. A second logical product extraction step that, the second
Area extraction step for extracting the area of each dark area of the image after the logical product extraction step, and a dark area having an area smaller than a predetermined area among the areas extracted by the area extraction step is determined to be a defect. A method for inspecting a surface defect, which comprises: (4) a plurality of respective dark portions have a Koritsu the dark portion and irradiation step Projects a light source <br/> Ru dark pattern to have a periphery to the inspected surface, the images of the specimen surface bright and dark pattern is projected An image capturing step of capturing into an image processing apparatus, a binarizing step of converting the image captured in the image capturing step into a bright and dark binary image, and an image binarized by the binarizing step. A contraction step of contracting a dark area of the image, a contour extraction step of making only the boundary between light and dark of the image after the contraction step bright, and an expansion step of expanding the bright area of the image after the contour extracting step. An inversion step of inverting the lightness and darkness of the image after the expansion step, a logical product extraction step of extracting a logical product of the image after the inversion step and the image after the binarization step, and the logical product An area extraction step of extracting the area of each dark region of the image after the extraction step; Of the area extracted by the extraction process, having a determination step of determining a dark region and the defect has a smaller area predetermined area, surface defect inspection method.

In the surface defect inspection methods (1) to (4) described above, first, in the irradiation step, a light source having a light and dark pattern having an isolated dark portion is projected on the surface to be inspected, and an image of the surface to be inspected on which the light and dark pattern is projected. Is taken into the image processing device, and the image is binarized into light and dark. When the image of the surface to be inspected on which the light and dark pattern is projected is captured in the image processing device, if a dark area that is not caused by the light and dark pattern is imaged in the light area, the surface to be inspected is not damaged. Possibly defective.
However, the isolated dark spots (dark areas) appearing around the dark areas in the image due to the dark portions of the light source of the light and dark patterns have a rough surface of the surface to be inspected, for example, a condition called yuzu skin. It appears because it is not a defect. for that reason,
In the surface defect inspection methods of (1) to (4) above, in a process after binarizing an image taken in an image processing device by a binarizing process, a dark spot due to surface roughness is referred to as a defect. Each process that is not judged is performed. In the surface defect inspection method of the above (1), in the binarized image, the area / center of gravity of each dark region is extracted in the area / center of gravity extraction step, and in the determination step, a dark area having a predetermined area or more is detected. A dark area located at a predetermined distance or more from each area is determined to be a defect. A dark region having an area equal to or larger than a predetermined area is a region caused by a dark portion of a light source having a light-dark pattern in each dark region of a binarized image, and is determined as a defect.
A dark area having a predetermined area or more (a dark area caused by a dark portion of a light source having a light-dark pattern), a dark area located at a predetermined distance or more, that is, a dark portion of a light source having a light-dark pattern The surface roughness of the surface to be inspected is not determined to be a defect because it is a dark area excluding the dark area and the surrounding area. In the surface defect inspection method of (2) above, 2
In the dilation process, the binarized image expands the dark areas (contracts the light areas), and thereby the dark areas caused by the dark areas of the light-dark pattern light source, and the surface to be inspected around the dark areas. The dark spots (dark areas) due to the surface roughness of are absorbed. Therefore, in the image after the expansion process, the dark area is either due to the dark portion of the light source of the light-dark pattern or due to the defect, and the dark point (dark area) due to the surface roughness of the surface to be inspected is Absent. In the determination step, among the dark areas of the image after the expansion step, the dark area having an area smaller than the predetermined area is determined to be a defect, but the image after the expansion step has a surface roughness of the surface to be inspected. Since there is no dark point (dark area) due to the, the surface roughness of the surface to be inspected is not judged as a defect. In the surface defect inspection method of the above (3), in the binarized image, the dark area is contracted (the light area is expanded) in the contraction step, and the dark spot caused by the surface roughness of the surface to be inspected. The (dark areas) and the dark areas due to the defects are absorbed in the light areas. Then, in the expansion step, each dark area after the contraction step is made larger than each dark area of the image immediately after binarization. Further, through the first inversion step, the first logical product extraction step, and the second inversion step, an image in which only the region near the boundary between light and dark of the image immediately after the binarization step is made dark is obtained. . Then, in the second logical product extraction step, the logical product of the image after the second inversion step and the image immediately after the binarization step is extracted, and the surface roughness of the surface to be inspected is compared with the image immediately after the binarization step. An image is obtained in which the dark spots (dark areas) caused by are removed. In the determination step, a dark area having an area smaller than a predetermined area out of the areas of the dark areas of the image after the second AND extraction step is determined to be a defect, but it is caused by the surface roughness of the surface to be inspected. Since the dark spots (dark regions) to be removed are removed in the second logical product extracting step, the surface roughness of the surface to be inspected is not judged as a defect. In the surface defect inspection method of the above (4), in the binarized image, the dark area is contracted (the bright area is expanded) in the contraction step, and the dark spot caused by the surface roughness of the surface to be inspected. The (dark areas) and the dark areas due to the defects are absorbed in the light areas. And
An image in which only the region near the boundary between light and dark of the image immediately after the binarization process is dark is obtained through the contour extraction process, the expansion process, and the inversion process. Then, in the logical product extraction process, the logical product of the image immediately after the inversion process and the image immediately after the binarization process is extracted, and the darkness due to the surface roughness of the surface to be inspected is compared with the image immediately after the binarization process. An image is obtained in which points (dark areas) are removed. In the determination step, a dark area having an area smaller than a predetermined area out of the areas of the dark areas of the image after the logical product extraction step is determined to be a defect, and a dark spot due to the surface roughness of the surface to be inspected. Since the (dark area) is removed in the logical product extraction step, the surface roughness of the surface to be inspected is not judged as a defect.

[0006]

1 is a schematic view of an apparatus for carrying out a surface defect inspection method common to all the embodiments of the present invention.
5 shows the procedure of the surface defect inspection method of the first embodiment of the present invention, FIGS. 3 and 4 show schematic diagrams of images obtained in the process of the method of the first embodiment of the present invention, and FIG. The flow chart shows the procedure of the surface defect inspection method of the second embodiment of the present invention, FIGS. 6 to 8 show schematic diagrams of images obtained in the process of the method of the second embodiment of the present invention, and the flow chart of FIG. Shows the procedure of the surface defect inspection method of the third embodiment of the present invention, FIGS. 10 to 14 show schematic diagrams of images obtained in the course of the method of the third embodiment of the present invention, and the flow chart of FIG. The procedure of the surface defect inspection method of the fourth embodiment of the present invention is shown, and FIGS. 16 and 17 are schematic views of images obtained in the process of the method of the fourth embodiment of the present invention.

First, a surface defect inspection apparatus that can be commonly used when performing the methods of all the embodiments of the present invention will be described with reference to FIG. The surface defect inspection apparatus processes an image with a camera 1 which is an apparatus for capturing an image of a surface to be inspected (including judgment).
Image processing device 2 and a light source 3 having a light and dark pattern having an isolated dark portion. The camera 1 has a convex lens. In the embodiment of the present invention, the light and dark pattern of the light source 3 is a black and white pattern. Each dark portion of black is, for example, a square,
All of them have the same size, and they are isolated in a white bright part and are regularly arranged at equal intervals.

The surface defect inspection of all the embodiments of the present invention can be carried out on an automobile body whose surface to be inspected is a mirror surface, for example, a metallic paint is applied. In the first to fourth embodiments of the present invention, an example of a method for detecting the presence / absence of a defect on the painted surface of an automobile body will be shown. The surface defect inspection is performed, for example, with the surface defect inspection device at a fixed position while the automobile body is conveyed by a conveyor.

First, steps of the surface defect inspection method common to all the embodiments of the present invention will be described with reference to FIG. (1) Irradiation step A black-and-white pattern light source having an isolated black portion (dark portion) is projected on a painted surface (inspected surface) of an automobile body.

(2) Image capturing step The image of the surface to be inspected on which the black-and-white pattern is projected is captured into the image processing apparatus via the camera equipped with the convex lens. An image (original image) captured by the image processing apparatus is displayed, for example, in FIG.
Shown in. A black (dark) point (also a black (dark) area) due to the orange peel of the surface to be inspected is copied around the black (dark) area due to the black portion of the black and white light source. The images captured by the image processing apparatus are sequentially processed by the image processing apparatus programmed with the following steps.

(3) Binarization step The value indicating the brightness of each pixel of the original image shown in FIG. 3, which is taken in by the image processing apparatus, is smoothed to obtain the value indicating the brightness of each pixel of the original image. And take the difference. Then, each pixel is binarized using a predetermined value indicating the brightness as a threshold value. Each pixel is set to either 1 or 0 by binarization. 1 is bright (white) and 0 is dark (black). In the image after the binarization process, a plurality of substantially quadrangular regions (regions of 0) in which 0 pixels are adjacent to each other and aggregated are arranged at equal intervals, and 1 pixel is arranged in other portions. Lined up.
However, in the following description, the image after the binarization process is
Alternatively, instead of writing 0, it is written white or black. Therefore, in the image after the binarization process, a plurality of black regions in which black (0) pixels are adjacent to each other are isolated in a white region in which white (1) pixels are adjacent to each other. It is an image. Around the isolated black region, black dots (which are also black regions, but are shown as black dots below) due to the orange peel of the surface to be inspected are scattered.

Since the steps after the above steps (1) to (3) are different for each embodiment,
The process performed after the process (3) will be described for each embodiment.

First, the method of the first embodiment of the present invention will be described with reference to FIGS. In the first embodiment of the present invention, the following steps (4) to (5) are performed after the steps (1) to (3).

(4) Area / Center of Gravity Extraction Step The position and center of gravity of the black (dark) region (isolated object) of the binarized image are extracted. A plurality of black areas are isolated in a white (bright) area. By extracting the center of gravity of each black area, the position of each black area can be grasped.

(5) Judging Step When the area of the isolated black region is equal to or larger than the predetermined area, it is judged that the black region (object) is caused by the black portion of the black-and-white light source. Further, when there is a black spot around a black region having an area equal to or larger than a predetermined area, it is attributed to the surface roughness of the surface to be inspected. Therefore, by setting the black area (object) that is determined to be a defect as a black area that is located at a position that is separated by a predetermined distance or more from a black area (object) that has an area equal to or larger than a predetermined area, The black dots due to the roughness are not determined as defects, and the surface roughness of the surface to be inspected is not determined as defects. More preferably, the surface roughness of the surface to be inspected is determined by determining that a black area having a predetermined distance or more from a black area having a predetermined area or more and having a predetermined area or more is a defect. Can be reliably prevented from being determined as a defect.

Next, the method of the second embodiment of the present invention will be described with reference to FIGS. In the second embodiment of the present invention, the following steps (6) to (8) are performed after the steps (1) to (3).

(6) Expansion step The black area resulting from the black portion of the black and white pattern of the binarized image (see FIG. 6) is expanded (see FIG. 7). Expanding a black area is also contracting a white area. The degree of expansion of the black area is determined by observing the black dots due to the surface roughness of the surface to be inspected, which are scattered around the black area of the binarized image apart from the black areas. It is only absorbed to each black area due to the black portion of As a result, the black spots caused by the surface roughness of the surface to be inspected are absorbed by the expanded (expanded) black region and become part of the black region.

(7) Area Extracting Step The area of each isolated black region of the image after the expanding step is extracted. In the area extraction step, first, isolated black areas are labeled in order as 1, 2, 3, ... And a histogram showing how many pixels having desired brightness are included in each labeled area. To create. The area of each labeled black region can be known from the number of pixels.

(8) Judgment process A black area having an area smaller than a predetermined area is judged as a defect.
From the size of the black and white pattern of the light source, the area of the black region due to the black portion of the light source of the image immediately after the binarization process can be determined in advance, and as a result, due to the defect related to the black portion of the light source of the image after the expansion process. The area of the black area that is not used can be determined in advance. Therefore, if there is a black area having an area smaller than the area (predetermined area) of the black area that can be determined in advance, it is determined as a defect. In the expansion step before the determination step, the black spots due to the surface roughness of the surface to be inspected are removed from the image, so the surface roughness of the surface to be inspected is not judged as a defect.

Next, the method of the third embodiment of the present invention will be described together with its operation with reference to FIGS.
In the third embodiment of the present invention, the following steps (10) to (17) are performed after the steps (1) to (3).

(10) Shrinkage step Shrink the black area of the binarized image (see FIG. 6) (see FIG. 10). Contracting a black area is also expanding a white area. The degree of shrinkage in the black area is 2
It is the extent to which the black areas due to the surface roughness of the surface to be inspected and the black areas due to the defects in the black areas of the binarized image are absorbed by the white areas. As a result, the black area of the image is only due to the black portion of the black and white light source,
The area of the region is smaller than the black region due to the black and white pattern immediately after the binarization process.

(11) Expansion Step The black area of the image, which has been contracted in the contraction step, is expanded (see FIG. 11). As for the degree of expansion, black dots appearing due to the surface roughness of the surface to be inspected, which are scattered around the black area due to the black portion of the black and white pattern of the light source in the image immediately after the binarization process (see FIG. 6). It will be within the range. This allows
The area of the black region is larger than the area immediately after the binarization process.

(12) First inversion step The black and white of each pixel in the image obtained by the expansion step is inverted. As a result, the white of each pixel in the image obtained by the expansion step becomes black and the black becomes white.

(13) First logical product extracting step The logical product (AND) of the image after the contracting step (see FIG. 10) and the image after the first inverting step is extracted (see FIG. 12).

(14) Second inversion step The black and white of each pixel in the image after the first AND extraction step is inverted. As a result, the white of each pixel in the image obtained by the first logical product extraction step becomes black and the black becomes white, and only near the boundary between the white area and the black area of the image immediately after the binarization step. A blackened image is obtained. The vicinity of the boundary between the white area and the black area immediately after the binarization step is a portion where black dots due to the surface roughness of the surface to be inspected are scattered.

(15) Second logical product extraction step The logical product (AND) of the image after the second inversion step (see FIG. 12) and the image immediately after the binarization step (see FIG. 6) is extracted. (See Figure 13). As a result, an image obtained by removing the black dots due to the surface roughness of the surface to be inspected from the image immediately after the binarization process (see FIG. 6) is obtained.

(16) Area extraction step The area of each black region of the image after the second logical product extraction step is
Each is extracted by the same method as the method shown in the second embodiment. In the image after the second logical product extraction step, there are black regions due to the black portion of the light source having a black-and-white pattern and black regions due to the defects when the surface to be inspected is defective, and all black The area of each area is extracted.

(17) Judging Step A black area having an area smaller than a predetermined area is judged as a defect (see FIG. 14). Since the area of the black region of the image immediately after the binarization process can be determined in advance from the size of the black part of the black and white pattern of the light source, it is related to the black part of the light source of the image after the second AND extraction process and is not caused by a defect. The area of the black area can be determined in advance. Therefore, a black area having an area smaller than the area (predetermined area) of the black area that can be determined in advance is determined as a defect. Since the black dots due to the surface roughness of the surface to be inspected are removed from the image in the second logical product extracting step, the surface roughness of the surface to be inspected is not judged as a defect in the judging step.

Next, the method of the fourth embodiment of the present invention will be described with reference to FIGS. In the fourth embodiment of the present invention, the following steps (18) to (24) are performed after the steps (1) to (3).

(18) Shrinkage step The black area of the binarized image (see FIG. 6) is shrunk (see FIG. 10). Contracting a black area is also expanding a white area. The degree of shrinkage in the black area is 2
In the black area of the binarized image, a black spot (black area) due to the surface roughness of the surface to be inspected and a black area due to the defect are absorbed to the white area. As a result, the black area of the image is only due to the black and white pattern of the light source, and the area of that area is made smaller than the black area due to the black portion of the light source of the black and white pattern immediately after the binarization process. .

(19) Outline Extracting Step Only the boundary between each isolated black area and white area of the image after the contraction step is white, and the other is black (see FIG. 16).

(20) Expansion process The white area of the image after the contour extraction process is expanded (see FIG. 12). As a result, the obtained white area is an area around the black area due to the black portion of the light source having a black-and-white pattern in the image immediately after the binarization process, in which a black dot due to the surface roughness of the surface to be inspected appears. is there.

(21) Inversion step The black and white of each pixel in the image after the expansion step is inverted. As a result, white of each pixel becomes black and black becomes white, and an image in which only the vicinity of the boundary between the white area and the black area in the image immediately after the binarization process (see FIG. 6) is the black area is displayed. can get. The obtained black area is an area around the black area due to the black portion of the light source having a black-and-white pattern in the image immediately after the binarization process, in which black dots due to the surface roughness of the surface to be inspected appear.

(22) Logical product extraction process Logical product (A of the image after the inversion process and the image immediately after the binarization process (A
ND) is extracted. As a result, an image (see FIG. 17) in which black dots due to the surface roughness of the surface to be inspected are removed from the image immediately after the binarization process is obtained.

(23) Area Extracting Step The area of each black region after the logical product extracting step is extracted by the same method as the method shown in the second embodiment. In the image after the logical product extraction process, there are black areas due to the black portion of the black and white light source and black areas due to the defect if the surface to be inspected is defective. Extract each.

(24) Judging Step A black area having an area smaller than a predetermined area is judged as a defect (see FIG. 14). From the size of the black portion of the black and white light source, the area of the black region that is not caused by the defect immediately after the binarization process can be determined in advance. The area of the region can be determined in advance. Therefore, a black area having an area smaller than the area (predetermined area) of the black area that can be determined in advance is determined as a defect. Since the black dots due to the surface roughness of the surface to be inspected are removed from the image in the logical product extraction step before the judgment step, the surface roughness of the surface to be inspected is not judged as a defect.

[0037]

According to the surface defect inspection method of claim 1,
After extracting the area and the center of gravity of each dark area of the binarized image in the area / centroid extraction step, the darkness at a position separated by a predetermined distance or more from each dark area having an area equal to or larger than the predetermined area in the determination step. Since the area is determined to be a defect, a dark point due to the surface roughness of the surface to be inspected, which appears within a predetermined distance from each dark area having a predetermined area or more, is not determined to be a defect. According to the surface defect inspection method of claim 2, by expanding the dark area of the binarized image in the expansion step, the dark point caused by the surface roughness of the surface to be inspected is absorbed in the dark area. Therefore, the dark area where the area is extracted in the area extraction step is the area excluding the dark spots (dark areas) due to the surface roughness, and the dark area due to the surface roughness of the surface to be inspected in the determination step. Is never considered defective. According to the surface defect inspection method of claim 3, in the shrinking step, the dark area of the binarized image is shrunk (the bright area is expanded), so that the dark area caused by the dark portion of the light-dark pattern. Dark spots (dark areas) caused by the surface roughness of the surface to be inspected around the
And the dark spot due to the defect is absorbed in the bright region, and the dark region becomes an image larger than the dark region of the image immediately after being binarized in the expansion process. Further, through the first inversion step, the first logical product extraction step, and the second inversion step, an image in which only the position near the boundary between light and dark of the image immediately after binarization is dark is obtained. To be Then, in the second logical product extraction step, the logical product of the image after the second inversion step and the image immediately after binarization is extracted, and the dark point (dark spot due to the surface roughness of the surface to be inspected Since the image in which the area) is removed is obtained, the surface roughness of the surface to be inspected is not determined to be a defect in the determination step performed based on the image after the second logical product extraction step. According to the surface defect inspection method of claim 4, in the shrinking step, the binarized image shrinks the dark area (expands the bright area) to thereby inspect the surface around the dark area. The dark spots due to the surface roughness and the dark spots due to defects are absorbed in the bright regions. Then, through the contour extraction process, the expansion process, and the inversion process, an image in which only the position near the boundary between the light and dark of the image immediately after binarization is dark is obtained. Further, in the logical product extracting step, the logical product of the image immediately after the inversion step and the image immediately after being binarized is extracted, and the dark spot (dark area) due to the surface roughness of the surface to be inspected is removed. Since an image is obtained, the surface roughness of the surface to be inspected is not determined to be a defect in the determination step performed based on the image after the logical product extraction step.

[Brief description of drawings]

FIG. 1 is a diagram showing an outline of an apparatus for performing a surface defect inspection method common to all embodiments of the present invention.

FIG. 2 is a flowchart showing a procedure of a surface defect inspection method according to the first embodiment of the present invention.

FIG. 3 is a schematic diagram of an image obtained in the process of the method according to the first embodiment of the present invention.

FIG. 4 is a schematic diagram of an image obtained in the process of the method of the first embodiment of the present invention.

FIG. 5 is a flowchart showing a procedure of a surface defect inspection method according to a second embodiment of the present invention.

FIG. 6 is a schematic diagram of an image obtained in the process of the method of the second embodiment of the present invention.

FIG. 7 is a schematic diagram of an image obtained in the process of the method of the second embodiment of the present invention.

FIG. 8 is a schematic diagram of an image obtained in the process of the method of the second embodiment of the present invention.

FIG. 9 is a flowchart showing a procedure of a surface defect inspection method according to a third embodiment of the present invention.

FIG. 10 is a schematic diagram of an image obtained in the process of the method according to the third embodiment of the present invention.

FIG. 11 is a schematic diagram of an image obtained in the process of the method of the third embodiment of the present invention.

FIG. 12 is a schematic diagram of an image obtained in the process of the method of the third embodiment of the present invention.

FIG. 13 is a schematic diagram of an image obtained in the process of the method of the third embodiment of the present invention.

FIG. 14 is a schematic diagram of an image obtained in the process of the method according to the third embodiment of the present invention.

FIG. 15 is a flow chart showing a procedure of a surface defect inspection method according to a fourth embodiment of the present invention.

FIG. 16 is a schematic diagram of an image obtained in the process of the method of the fourth embodiment of the present invention.

FIG. 17 is a schematic diagram of an image obtained in the process of the method of the fourth embodiment of the present invention.

[Explanation of symbols]

1 camera 2 Image processing device 3 light sources

─────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-9-318338 (JP, A) JP-A-8-86633 (JP, A) JP-A-5-91411 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) G01B 11/00-11/30 G01N 21/84-21/958 G06T 1/00 G06T 7/00 G06T 7/60

Claims (4)

(57) [Claims] [Claim 1 wherein each dark portion have a plurality of Koritsu the dark portion around
And Projects irradiation step the source of light and dark patterns on the inspected surface to have a bright and process the image captured in step capture the image capture image capture images of the inspected surface bright and dark pattern is projected to the image processing apparatus And a binarization step of darkly binarizing the image, and an area / centroid extraction step of extracting the area and centroid of each dark region of the binarized image by the binarization step, respectively. Of the areas extracted by the area / centroid extraction process,
A surface defect inspection method, comprising: a dark area located at a predetermined distance or more from each dark area having an area equal to or larger than a predetermined area to be a defect. Wherein each dark portion have a plurality of Koritsu the dark portion around
And Projects irradiation step the source of light and dark patterns on the inspected surface to have a bright and process the image captured in step capture the image capture image capture images of the inspected surface bright and dark pattern is projected to the image processing apparatus And a binarization step of making a dark binarized image, an expansion step of expanding a dark area of the image binarized by the binarization step, and a dark area of the dark area after the expansion step. A surface defect inspection method comprising: an area extraction step of extracting each area; and a determination step of determining a dark region having an area smaller than a predetermined area as a defect among the areas extracted by the area extraction step. Wherein each dark portion have a plurality of Koritsu the dark portion around
And Projects irradiation step the source of light and dark patterns on the inspected surface to have a bright and process the image captured in step capture the image capture image capture images of the inspected surface bright and dark pattern is projected to the image processing apparatus And a binarization step for making a dark binarized image, a shrinking step for shrinking a dark area of the image binarized by the binarizing step, and a dark area of the image after the shrinking step. An expansion process of expanding the image, a first inversion process of inverting the lightness and darkness of the image after the expansion process, a logical product of the image after the contraction process and the image after the first inversion process A first logical product extracting step, a second inverting step of inverting the light and darkness of the image after the first logical product extracting step, an image after the second inverting step, and the binarization A second logical product extracting step for extracting a logical product with the image after the process, and the second theory An area extraction step of extracting the area of each dark area of the image after the product extraction step, and a determination is made that a dark area having an area smaller than a predetermined area among the areas extracted by the area extraction step is determined to be a defect. A surface defect inspection method comprising: Wherein each dark portion have a plurality of Koritsu the dark portion around
And Projects irradiation step the source of light and dark patterns on the inspected surface to have a bright and process the image captured in step capture the image capture image capture images of the inspected surface bright and dark pattern is projected to the image processing apparatus And a darkening binarized image, a shrinking process of shrinking a dark region of the binarized image by the binarizing process, and a bright and dark image of the image after the shrinking process. Contour extraction step for making only the boundary of the image bright, an expansion step for expanding the bright area of the image after the contour extraction step, an inversion step for inverting the light and dark of the image after the expansion step, and the inversion step A logical product extracting step for extracting a logical product of the subsequent image and the image after the binarizing step, and an area extracting step for extracting the area of each dark region of the image after the logical product extracting step, respectively. Of the areas extracted by the area extraction step, Having a determination step of determining a dark region and the defect has a smaller area area, surface defect inspection method.