JPH0682390A - Method and apparatus for inspecting surface defect - Google Patents

Abstract

PURPOSE:To accurately detect a delicate color irregular surface defect by subtracting a specific offset value from two or more color components of the image of the surface defect and adding the respective color components to emphasize a minute defect signal. CONSTITUTION:The surface defect 12 of matter 10 to be inspected and the peripheral part thereof are taken by a unidimensional imaging device 22 such is a three-color television camera and the video signal of the device 22 is converted to a digital signal by an A/D converter 24 to be sent out to an image processor 26. In the processor 26, the video signal is accumulated in a time series manner at every one line to make an image two-dimensional and the value obtained, for example, by substracting 30 from the brightness average value of the respective color (red, blue, green) components of the image is subtracted from the images of the respective color components as the offset values of the respective components and the images of the respective color components from which the offset values are subtracted are added to form a variable- density image and the signal of the weak defect 12 is emphasized. The output of the processor 26 is set to an operator 28 and the presence of the defect 12 is judged from a binarized area to be outputted and the defect 12 is displayed on an image monitor 30 as an image.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a surface defect inspection method and apparatus for detecting surface defects such as cold-rolled steel sheet, and in particular, processing a surface defect image picked up by a television camera to obtain a delicate color. The present invention relates to a surface defect inspection method and apparatus capable of accurately detecting uneven surface defects and the like.

[0002]

2. Description of the Related Art Generally, as a method for detecting surface defects on a cold-rolled steel sheet or the like, spot-like laser light is applied to the surface of the cold-rolled steel sheet or the like, as disclosed in JP-A-62-75234. There is known a surface defect detection method using a laser scattering method or a laser diffraction method, in which a surface defect is detected by irradiating and utilizing a scattering or diffraction phenomenon of laser light generated by the unevenness of the surface.

On the other hand, as a method of detecting a surface defect of an object to be inspected from image data obtained by photographing the surface of the object to be inspected by a television camera, for example, as disclosed in JP-A-56-77704, a threshold value is used. The binarization method is used, and when the image obtained by imaging the surface of the object to be inspected with a television camera shows a brightness higher than a set value or lower than the set value, the surface is regarded as a defect. It is decided.

[0004]

However, in the above laser scattering method and laser diffraction method, it is not possible to detect color irregular surface defects, and it is difficult to detect defects due to subtle changes in surface roughness. There was a problem.

Further, in the binarization method using a television camera, it is difficult to distinguish minute defects from diffraction and scattered light due to surface roughness, and when binarization is performed with a constant threshold value, There is a problem that a large difference occurs in the binarization range when a change in signal intensity occurs due to a change in reflectance of the surface of the object to be inspected or a change in illumination light intensity.

The present invention has been made to solve the above-mentioned conventional problems, and it is possible to accurately detect a subtle uneven color-like surface defect by processing a surface defect image picked up by a television camera. An object of the present invention is to provide a method and apparatus for detecting surface defects.

[0007]

SUMMARY OF THE INVENTION The present invention is a surface defect inspection method for detecting a surface defect of an object to be inspected from image data obtained by imaging the surface, and an image pickup device capable of picking up two or more color components. By solving the above problem, a specific offset value is subtracted from each of the photographed color components, and each color component is added to create an imaging screen that emphasizes a minute defect signal. .

Further, by obtaining the offset value from the average value of each color component in the image, the influence of the change of the reflectance of the surface and the change of the intensity of the illumination light is reduced.

Further, when the image pickup screen created as described above is binarized with a preset value and the binarized area exceeds a preset discrimination threshold, a defect is found in the photographed image. It is determined that it exists.

Similarly, the present invention has, in a surface defect inspection apparatus, a light source for irradiating an object to be inspected and a function of picking up an image of the surface of the object to be inspected and transmitting an image signal as a video signal. , An analog / digital signal converter having a function of converting a video image signal into a digital image signal, and an image data memory for storing image data, and an average value of luminance in an image. , A specific offset value is subtracted, the values of the respective color components are added, and an image processing device having a function of binarizing, a discriminating device that determines the presence or absence of a defect from the binarized area, and the determination result The above-mentioned problem is achieved by including an output device for outputting.

[0011]

[Function] Up to now, in the inspection of steel plate surface defects by image pick-up by a television camera, diffraction, scattered light due to the surface roughness of the steel plate surface,
Furthermore, it was difficult to identify a weak irregular color defect due to a change in reflectance of the steel plate surface and a change in illumination light intensity.

Therefore, in the present invention, the average value of the brightness of each color component of the image data in one image picked up by an image pickup device capable of picking up two or more color components is obtained. A value slightly smaller than this average value is set as the offset value of each component and subtracted from the image of each color component. By adding the images obtained by subtracting the offset value from each color component and displaying it as a grayscale image, the signal of the subtle color irregularity defect is emphasized, and the scattered light on the steel plate surface and the electronic noise of the image sensor are added together. By doing so, they are averaged, and the SN ratio of the irregular color defect is improved.

Further, by subtracting the offset value obtained from the average value, it is possible to correct the changes in the reflectance and the illumination light intensity on the surface of the steel sheet.

Further, the grayscale image is binarized by a preset threshold value to binarize the defective portion, and the presence or absence of the defect can be determined by the binarized area.

[0015]

Embodiments of the present invention using a one-dimensional image pickup device having three one-dimensional image pickup elements for detecting three color components will be described in detail below with reference to the drawings.

FIG. 1 is an explanatory diagram showing the configuration of this embodiment. In this figure, the object to be inspected 10 made of, for example, a steel plate
There is a surface defect 12 in the surface of the inspection object 10, and the illumination light source 20 irradiates the inspection object 10 with light so as to illuminate the periphery of the surface defect 12.

22 is a color component of two or more colors (in the embodiment,
A one-dimensional image pickup device (for example, a three-plate three-color television camera) having two or more (three in the embodiment) one-dimensional image pickup device capable of picking up three components of red, green, and blue. The surface defect 12 of the object 10 and its peripheral portion are imaged. The output of the one-dimensional imaging device 22 is transmitted as a video signal,
After being A / D converted by the analog / digital (A / D) converter 24, it is sent to the image processor 26 for image processing.

In the image processing device 26, the video signal from the one-dimensional image pickup device 22 is accumulated line by line in synchronism with the movement pitch of the object 10 to be inspected, and the video signal for each line is sequentially arranged in time series. By accumulating, the image of the inspection object 10 is two-dimensionalized.

The output of the image processing unit 26 is output by the arithmetic unit 28.
Then, the calculation result is output from the binarized area to determine the presence or absence of a surface defect, and the calculation result is output via the calculation result output device 30. The output of the arithmetic unit 28 is also sent to the image monitor 30 to display an image showing the surface defect 12 clearly.

FIG. 2 shows red (wavelength range 600 to 800 n
m), green (wavelength range 500 to 600 nm), and blue (wavelength range 400 to 500 nm), which are imaged by a one-dimensional image pickup device 22 capable of picking up three color components. is there. A of the blue component of this image
The luminance distribution between −b is shown in FIG. Table 1 below shows the average value of the brightness of the image of each color component in these main images.

[0021]

[Table 1]

When the value obtained by subtracting 30 from the average value is set as the offset value of each component and subtracted from the image of each color component, the brightness distribution shown in FIG. 4 is obtained. In the figure,
(R) shows a red component, (G) shows a green component, and (B) shows a blue component.

When the images of the respective color components from which these offset values have been subtracted are added together as a grayscale image, the image shown in FIG. 5 is obtained, and its luminance distribution is shown in FIG. In this way, the signal of the weak defect portion is emphasized by a series of processes, and the reflectance of the steel plate surface and the change of the illumination light are corrected.

Further, when the binarization process is performed with the threshold value, the image shown in FIG. 7 is obtained, and the defective portion is extracted.

The above series of image processing is performed by the image processing device 26 of FIG.

Calculating the binarized area from the image of FIG.
The area ratio with respect to the display screen is about 6%, and defect determination can be performed based on the area ratio.

This defect determination is performed by the arithmetic unit 28, and the determination result is sent to the arithmetic result output unit 30 for printing or the like. The determination result is also sent to the image monitor 32 and displayed on the monitor 32.

In the above embodiment, a three-plate type three-color television camera was used as the image pickup device, but the type of the image pickup device is not limited to this, and two or more color components can be picked up. Any other device may be used as long as it includes one or more one-dimensional image pickup devices.

The image pickup device is not limited to the one-dimensional image pickup device, and a two-dimensional image pickup device capable of separating each color component for image pickup may be used. In the configuration using the two-dimensional image sensor, in the image processing device 26, as in the configuration using the one-dimensional image sensor, the video signal is accumulated line by line in synchronization with the movement pitch of the inspected object 10. It is not necessary to form a two-dimensional image, and the two-dimensional video signal from the image sensor can be directly processed.

Further, the object to be inspected is not limited to the surface-treated steel plate, but can be similarly applied to general steel plates and other metal plates.

[0031]

As described above, according to the present invention,
By processing a surface defect image captured by an image capturing device capable of capturing color components of two or more colors, it is possible to accurately detect subtle irregular color irregular surface defects that have been difficult to detect up to now. .

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing an example of an image including a surface defect imaged by the image pickup apparatus of the embodiment.

FIG. 3 is a diagram showing a luminance profile of a blue component between a and b in FIG.

FIG. 4 is a diagram showing a luminance profile obtained by subtracting a specific offset value from an image of each color component in the embodiment.

FIG. 5 is a diagram showing an image in which each color component from which an offset value has been subtracted is added together in the same manner.

FIG. 6 is a diagram showing a luminance profile between a and b in FIG.

FIG. 7 is a diagram showing an image obtained by performing binarization processing on FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Inspected object 12 ... Surface defect 20 ... Illumination light source 22 ... One-dimensional imaging device 24 ... A / D conversion device 26 ... Image processing device 28 ... Calculation device 30 ... Calculation result output device 32 ... Image monitor

Claims (4)

[Claims] 1. A surface defect inspection method for detecting a surface defect of an object to be inspected from image data obtained by imaging a surface, each color imaged by using an image pickup device capable of picking up two or more color components. A surface defect inspection method characterized by creating an imaging screen that emphasizes a minute defect signal by subtracting a specific offset value from the components and adding up each color component. 2. The offset value according to claim 1,
A surface defect inspection method characterized by reducing the influence of a change in the reflectance of the surface or a change in the intensity of illumination light by obtaining the average value of each color component in the image. 3. The image pickup screen created according to claim 1, binarized by a preset value, and when the binarized area exceeds a preset discrimination threshold, A surface defect inspection method characterized by determining that a defect exists. 4. A light source for irradiating an object to be inspected, an image pickup apparatus having a function of picking up an image of the surface of the object to be inspected and transmitting an image signal as a video signal, and a video image. An analog / digital signal converter having a function of converting a signal into a digital image signal, and an image data memory for storing image data, and calculating an average value of luminance in an image to subtract a specific offset value, An image processing device having a function of binarizing by adding the values of respective color components, a discriminating device for discriminating the presence or absence of a defect from the binarized area, and an output device for outputting the discrimination result. And surface defect inspection equipment.