JPH01206242A - Method and apparatus of checking curved surface - Google Patents

Abstract

PURPOSE:To automate a checking operation in a non-contact manner and with high precision, by applying a light so that an inclined linear light image be formed on a curved surface to be checked up, and by determining the presence or absence of a defect depending upon whether this linear light image is deformed or not. CONSTITUTION:A linear light source 12 is installed in inclination at a prescribed angle to the rotating axis of an object 10 of checking. Therefore a reflection light image 11 inclined to the rotating axis is formed on the surface of the object 10 of checking. The object 10 is set on a rotary stage 13 which is rotated by a rotating device 16, and the whole surface of the object is checked up during one rotation. A two-dimensional image pickup device 16 is disposed in the front of the object 10 of checking and senses a reflected light image from the object 10. A defect determination circuit 34 judges the size of a defect from an image stored in a defect image memory 32 and determines whether it is a defect to be removed or not.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a curved surface inspection method and apparatus for inspecting defects on a curved surface to be inspected, such as the surface of a cylindrical object.

[Conventional technology]

Beer and juice cans with defects such as dents and scratches must be inspected at the shipping stage to prevent them from being put on the market.

In addition, when shipping or receiving empty cans for filling beer, juice, etc., it is necessary to inspect the cans for defects such as 6 dents or scratches on the surface of cylindrical objects such as cans. Conventionally, the presence or absence of defects has been visually inspected by workers.

[Problem to be solved by the invention]

However, in visual inspection, which is based on human senses, the test results vary depending on the level of skill and fatigue of the operator, and there are problems with the reliability of the test.

The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a curved surface inspection method and apparatus that can automatically perform highly accurate and reliable inspections.

[Means for Solving the Problems] The above object is to provide a curved surface inspection method for inspecting defects on a curved surface to be inspected, in which a linear optical image tilted with respect to the axis of rotation of the curved surface to be inspected is placed on the curved surface to be inspected. irradiating the curved surface to be inspected so that the curved surface to be inspected is formed, capturing an image of the curved surface to be inspected including the linear light image, and based on whether or not the linear light image in the image of the curved surface to be inspected is deformed. This is achieved by a curved surface inspection method characterized in that the presence or absence of a defect is determined on the curved surface to be inspected.

The above object is to provide a curved surface inspection apparatus for inspecting defects on a curved surface to be inspected so that a linear optical image tilted with respect to a rotation axis of the curved surface to be inspected is formed on the curved surface to be inspected. an irradiation means for irradiating the image of the curved surface to be inspected including the linear light image; and an imaging means for capturing the image of the curved surface to be inspected including the linear light image; This is achieved by a curved surface inspection apparatus characterized by comprising a determining means for determining the presence or absence of a defect on the inspected curved surface.

[For production]

Since the present invention is configured as described above, light is irradiated so that an inclined linear optical image is formed on the curved surface to be inspected,
The presence or absence of a defect is determined based on whether or not this linear optical image is deformed.

〔Example〕

Hereinafter, the present invention will be explained based on an illustrated embodiment.

FIG. 1 shows a curved surface inspection apparatus according to an embodiment of the present invention.

In this embodiment, a case where a cylindrical aluminum can is inspected as the object to be inspected 10 will be described as an example.

A linear light source 12 is arranged to face the object 10 to be inspected. This linear light source 12 is installed at a predetermined angle with respect to the rotation axis of the object 10 to be inspected. Therefore, a reflected light image 11 tilted with respect to the rotation axis is formed on the surface of the object 10 to be inspected.

The linear light source 12 forms a linear light image on the surface of the object to be inspected. For example, it may be a linear light source such as a line filament or a column photodiode, or a linear slit may be provided in front of the diffused light source. Good too.

The object to be inspected 10 is placed on a rotary table 13, and the rotary table 13 is rotated by a rotating device 14, so that the rotary table 13
While rotating, the entire surface of the object to be inspected 10 is inspected. The two-dimensional imaging device 16 is placed in front of the object to be inspected 10 and captures a reflected light image of the object to be inspected 10 .

In order to clarify the arrangement relationship between the object to be inspected 10, the linear light source 12, and the two-dimensional imaging device 16, FIG. 2 shows a plan view of the curved surface inspection device viewed from above. 12 and the two-dimensional imaging device 16 are installed on the mount 15, the linear light source 12 is installed near the object to be inspected 10, and the two-dimensional imaging device 16 captures the reflected image of the object to be inspected 10. It is installed at a predetermined distance so that it can be seen.

The image of the object to be inspected 10 captured by the two-dimensional imaging device 16 is stored in a multilevel image memory 18 .

FIG. 3(a) shows a specific example of an image of the object 10 to be inspected.

In the image of FIG. 3(a), since the reflected light @11 of the object to be inspected 10 is straight, it can be seen that there is no defect on the surface of the object to be inspected 10.

The inspection area determination circuit 20 is for determining an inspection area 21, which is an area for actually determining the presence or absence of a defect, from the image of the object to be inspected 10 stored in the image memory 18. This inspection area 21 is predetermined so that the image portion of the object to be inspected 10 is contained therein, as shown by the dashed line in FIG. 3(b). In FIG. 3(b), since the object to be inspected 10 has a dent, the reflected light image 11 is deformed 11a.

The linear light image extraction circuit 22 extracts the inspection area 2 of the image memory 18.
A linear light image is extracted from the position of maximum brightness when looking inside 1 in the horizontal direction. The inspection area 21 is scanned in the horizontal direction, the brightness of dots separated by, for example, four dots on the scanning line is compared, and the address of the dot with the higher brightness is stored in the address register 2.
Store in 4.

When this process is performed on a scanning line from one end to the other, the address register 24 will finally contain a dot with the maximum brightness on that scanning line, that is, a linear optical image, as shown in FIG. 3(C). The location address will be stored in . When the above-described processing is performed for each scanning line in the horizontal direction of the inspection area 21, an address indicating the position of the reflected light image 11 of the inspection area 21 is stored in the address register 24.

Since the reflected light image 11 is formed obliquely on the object to be inspected 10, it is also oblique within the inspection area 21. The angle correction circuit 26 corrects the angle of the image in the image memory 18 so that the oblique reflected light image 11 becomes vertical, and stores the corrected image in the image memory 28. That is, the image angle correction circuit 26 first calculates the inclination angle of the reflected light image 11 based on the address of the maximum brightness dot stored in the address register 24. The image is corrected so that the top of the reflected light image 11 is located at the center of the upper part of the image memory 28, and the lower part of the reflected light image 11 is located at the center of the lower part of the image memory 28, as shown in FIG. 3(d).
) and stored in the binary image memory 28. This modification is done by converting the address when reading the image from the image memory 18.

Note that if the angle of the reflected light image 11 does not change much depending on the object to be inspected 10, it may be corrected to a predetermined angle.

Then, set the maximum brightness dot to "1" and set the other dots to "1".
When the pattern of "1" and "0" is formed, the reflected light image 11 becomes vertical as shown in FIG. 3(d) and is converted from multi-value to binary.

The defective point detection circuit 30 detects defective points from the image stored in the image memory 28. The method of detecting a defective point by this defective point detection circuit 30 is as follows.

First, horizontal lines drawn a predetermined distance in the image memory 28 are compared. Compare the corresponding dots on the horizontal lines. As a result of the comparison, if the dot values are the same, it is set as "0", otherwise it is set as "1".
0 image memory 2 where the dots remaining as "1" are defective points
Perform this process starting from the top line of 8, and get the result shown in Figure 3(e).
A processed image showing a defective point 11a- as shown in FIG. 1 is stored in the same image memory 28.

The defect detection circuit 30 further compresses the image in the image memory 28 shown in FIG. 3(e) into one-dimensional column data.
The defect image is stored in the defect image memory 32. Data compression is performed by the defect detection circuit 30 using, for example, the following algorithm.

■ If there is a predetermined number or more of consecutive "1" dots in a horizontal line in the image in the image memory 28 in FIG. Make it.

(2) Count the number of "1" dots in horizontal lines in the image in the image memory 28, and if the total value exceeds a predetermined number, it is determined as a defective line, and the bit of the column data corresponding to that line is set to "1".

When the processing for all lines is completed by the above-described method, one column of column data in the defective image memory 32 including the defective point 11'' has been generated, which includes the defective point stored in the image memory 28. This means that the compression of the image has been completed.In other words, this one column of data represents the state of the surface of the object to be inspected 10 on which the reflected light image 11 has been formed.

Therefore, if there is a defect on the surface of the object to be inspected 10, that portion will be represented as "1".

The above processing is processing of an image of the object to be inspected 10 in a stationary state at a certain rotational position, or in other words, processing of the reflected light image 11 on the object to be inspected 10. By rotating the object 10 to be inspected by the rotating device 14 and performing the above processing at fixed time intervals in synchronization with this rotation, an image showing defective points over the entire circumference of the object 10 to be inspected is obtained as shown in FIG.
The image is stored in the defective image memory 32 as shown in f). That is, the image stored in the defect image memory 32 corresponds to an image obtained by flattening the entire circumference of the object to be inspected 10. “1” in the defect image memory 32 is the object to be inspected 1
This corresponds approximately to the position and size of defects such as unevenness and scratches on the 0 surface.

The defect determination circuit 34 determines the size of the defect from the image stored in the defect image memory 32, and determines whether the defect should be eliminated. Determination is made by scanning the image in the defective image memory 32 in both the horizontal and vertical directions. Note that in this determination, since independent defects may appear in the upper and lower regions as shown in FIG. It is desirable to
It is desirable to provide an overlapping portion in each area.

The defect determination circuit 34 determines defects using, for example, the following algorithm.

■ Linear defects are judged by their length. That is,
A defect is determined when N1 or more "1"s are consecutive in one vertical or horizontal scanning line, or when the total number of "1"s is N2 or more.

■ Defects on the surface are determined by their area. That is,
If N3 or more consecutive vertical or horizontal scanning lines are consecutive, N4 or more consecutive lines are determined to be defective.

When the presence or absence of a defect is determined by such an algorithm, the defect determination circuit 34 outputs the determination result. The processing method for this determination result differs depending on the system in which the curved surface inspection apparatus of this embodiment is incorporated.

For example, if the curved surface inspection device of this embodiment is installed in a production line that fills cans with contents such as beer or juice, if a determination result indicating that there is a defect is output, an alarm lamp will be turned on or the production line will be Eliminate defective cans.

Next, the operation will be explained.

A zero-rotation device 1 that places an object to be inspected 10 on a rotary table 13 and rotates the rotary table 13 with a rotation device 14.
The two-dimensional imaging device 16 starts imaging the object to be inspected 10 in response to the timing signal from 4. A still image of the object to be inspected 10 (FIG. 3(b)) captured at a predetermined timing is stored in the image memory 18. Within the inspection area determined by the inspection area determination circuit 20, the linear light image extraction circuit 22 extracts the reflected light image 11 of the linear light source 12, and stores the address of the extracted reflected light 111 in the address register 24. (FIG. 3(C)), the image angle correction circuit 26 calculates the inclination angle of the reflected light image 11 of the object to be inspected 10 based on the address of the reflected light image 11 stored in the address register 24,
The reflected light image 11 is corrected so that it is vertically straight, and is simultaneously binarized and stored in the image memory 28 (FIG. 3(d)).The defect detection circuit 30 stores the image in the image memory 28. Defect points are detected from the image and stored as one dot line in the defect image memory 32 (FIG. 3(e)).
.

The above-described process is performed for the next rotational position of the object to be inspected 10, the next row of dot lines is stored in the defect image memory 32, and these processes are repeated to record the defect over the entire circumference of the object to be inspected 10 in the defect image memory 32. Image (Figure 3(f))
Store.

When the storage of the defect image of the entire circumference in the defect image memory 32 is completed, the presence or absence of a defect is determined by the defect determination circuit 34,
The judgment result is output.

In this manner, according to this embodiment, defects are detected from images captured by the two-dimensional imaging device, so defects can be detected without contact.

Further, since the linear optical image is extracted from a still image of the object to be inspected captured by the two-dimensional imaging device, defects can be easily detected even if the positional accuracy of the object to be inspected is low. Furthermore, since the reflected light image of the linear light source is formed on the object to be inspected in an inclined manner, defects in the object to be inspected are relatively amplified and the defects can be easily detected. Also,
In the curved surface inspection apparatus of this embodiment, since the only mechanical part is around the rotating device 14, the failure rate is low and maintenance is easy.

A curved surface inspection apparatus according to another embodiment of the present invention is shown in FIG. Components that are the same as those in the above embodiment are given the same reference numerals and explanations will be omitted. This example uses two linear light sources 121 and 122.
It is characterized in that the object to be inspected 10 is irradiated with. The linear light sources 121 and 122 are arranged symmetrically and opposite the object to be inspected 10 . The linear light sources 121 and 122 are arranged so that their reflected light images 111 and 112 are oblique to the rotation axis of the object 10 to be inspected. Moreover, as shown in FIG. 5, the reflected light image 111 of the linear light source 121 and the linear light source 1
The positions of the 22 reflected light images 112 are adjusted so that they are not in the same direction but at a predetermined angle.

The image of the object to be inspected 10 including the reflected light images 111 and 112 is 2
The image is captured by the dimensional imager 16 and processed in the same manner as in the previous embodiment. However, as shown in FIG. 5, since there are two reflected light images 111 and 112 in one image, the inspection area determination circuit 20 divides the image into two inspection areas, left and right.
A linear optical image is extracted for each inspection area to extract defective points. The column dot lines indicating the extracted defect points are each stored in the defect image memory 32 as separate defect images, and the defect determination circuit 34 is used for each defect image.
The presence or absence of a defect is determined by this.

If it is determined that a defect exists in any defective image, the determination result will be "defect present".

In this manner, in this embodiment, defects are detected using reflected light images of two linear light sources that form a predetermined angle with each other, so that no defects are missed.

If a defect happens to be perpendicular to the linear light source, it may not be detected as a defect, but according to this embodiment, even if the defect is perpendicular to one reflected light image and cannot be detected, it can be detected in the other reflected light image. This is because it can be detected because it forms an angle.

In the above-described embodiment, if the printed pattern 10a is applied to the surface of the object to be inspected 10, there is a possibility that the printed i*i0a may be mistakenly recognized as a defect. In order to prevent this, the processing algorithm in the linear optical image extraction circuit 22 is set as follows by utilizing the characteristics of the deformation 11b of the linear optical image by the printed pattern 10a. The optical image becomes a cut-off portion 11b as shown in FIG. However, this broken part 11b is short and does not continue for a long time like a dent or a scratch. Therefore, when extracting a linear optical image by the linear optical image extraction circuit 22, a short When the addresses stored in the address register 24 are scattered due to a break, the addresses are made to match the previous and subsequent addresses.

The present invention is not limited to the above embodiments, and various modifications are possible.

For example, in the above embodiments, beer and juice cans were inspected for defects, but cans with rotating bodies such as cylindrical bodies and conical bodies may also be inspected.
Also, instead of a can, a pipe, transparent bottle, or opaque field may be used.
Furthermore, even when a flat plate such as a metal sheet or film is wound, a curved surface is formed in some parts, so defects in the flat plate can be detected by irradiating this part with a linear light source.

In short, defects on any surface can be detected as long as the surface to be inspected is a curved surface.

In addition, defects may be detected by irradiating the surface of the object to be inspected with three or more linear light sources. Note that the defects that can be detected are not limited to scratches and irregularities, but also chips, cracks, bruises, and attached foreign objects. It is possible to detect any defects such as , poor shaping, etc.

Furthermore, if it is necessary to speed up the defect determination process, multiple systems of the configuration shown in FIG. 1 may be provided to perform parallel processing.

〔Effect of the invention〕

As described above, according to the present invention, even if the positioning accuracy of the curved surface to be inspected is not strict, highly accurate inspection can be automatically performed in a non-contact manner.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram of a curved surface inspection device according to an embodiment of the present invention, FIG. 2 is a plan view of the curved surface inspection device, FIG. 3 is a diagram for explaining the inspection process of the curved surface inspection device, and FIG. The figure is a plan view of a curved surface inspection device according to another embodiment of the present invention, FIG. 5 is a diagram for explaining the inspection process of the same curved surface inspection device, and FIG. 6 is an object to be inspected in a can with a printed pattern. FIG. 2 is a diagram showing a reflected light image of FIG. 10... Object to be inspected, 11... Reflected light image, 12...
・Linear light source, 13... rotating table, 14... rotating device,
15... Frame, 16... Two-dimensional imaging device, 18...
- Image memory, 20... Inspection area determination circuit, 22...
- Linear optical image extraction circuit, 24...address register, 2
6... Image angle correction circuit, 28... Image memory, 30
... Defect point detection circuit, 32 ... Defect image memory, 3
4...Defect determination circuit.

Claims (1)

[Claims] 1. In a curved surface inspection method for inspecting defects on a curved surface to be inspected, a linear optical image tilted with respect to a rotation axis of the curved surface to be inspected is formed on the curved surface to be inspected. irradiating the curved surface to be inspected, capturing an image of the curved surface to be inspected including the linear light image, and determining the curved surface to be inspected based on whether or not the linear light image in the curved surface to be inspected image is deformed; A curved surface inspection method characterized by determining the presence or absence of a defect. 2. The curved surface inspection method according to claim 1, wherein the curved surface to be inspected is irradiated so that a plurality of linear light images having different inclination angles are formed on the curved surface to be inspected. 3. In a curved surface inspection device for inspecting defects on a curved surface to be inspected, the curved surface to be inspected is irradiated so that a linear light image tilted with respect to the rotation axis of the curved surface to be inspected is formed on the curved surface to be inspected. irradiation means for capturing the image of the curved surface to be inspected including the linear light image; and imaging means for capturing the image of the curved surface to be inspected including the linear light image; 1. A curved surface inspection device comprising: determination means for determining the presence or absence of a defect. 4. The apparatus according to claim 3, wherein the irradiation means irradiates the curved surface to be inspected so that a plurality of linear light images having mutually different inclination angles are formed on the curved surface to be inspected. Curved surface inspection equipment.