JP2021099285A - Color unevenness inspection device and color unevenness inspection method - Google Patents

Abstract

To provide a device for evaluating the unevenness in color of a test object that can easily be recognized and quantified, the device being easily adoptable for various test objects such as colorized stainless steel, etc.SOLUTION: Provided is a color unevenness inspection device 100 comprising: image detection means 40 for detecting, for each pixel, a color component from an image of reflected and scattered light of irradiation light in the inspection range of a test object 21 and having a light source 11, a half mirror 13 for reflecting at least a portion of light irradiated from the light source 11 to the inspection face of the test object 21 and an imaging unit 30 for imaging light reflected or scattered from the inspection face and having passed through the half mirror 13 and thereby imaging an image in the inspection range of the inspection face; color system conversion means for converting, for each pixel, the color component into a color system component; reference value setting means for finding a reference value in the inspection range from a spatial distribution that uses at least two characteristic values of the color system component; and Nx calculation means for calculating, for each pixel in the inspection range of the image, a norm Nx in the spatial distribution from the reference value to the pixel.SELECTED DRAWING: Figure 1

Description

The present invention relates to a color unevenness inspection device and a color unevenness inspection method for color-treated stainless steel and the like.

Appearance inspection such as surface condition is one of the important inspection items in product evaluation. As a means for inspecting the surface condition of the object to be inspected, many inspection techniques utilizing various characteristics such as light reflection, scattering, absorption, polarization, and color have been disclosed and put into practical use. Since the characteristics of light vary depending on the surface condition of the object to be inspected, it is necessary to irradiate and detect the inspection light by grasping the mathematically analyzable optical behavior of light according to the surface condition. Become. For example, Patent Document 1 evaluates a state of unevenness and unevenness of distribution of unevenness on an object surface, and discloses, for example, an apparatus for evaluating gloss unevenness and the like.

In the visual inspection, evaluation of color distribution (color unevenness) may be required. Patent Document 2 is an apparatus for inspecting the color tone of an inspection object by image processing. After defining an inspection area for an input image of the inspection object input by an image input means, a predetermined number of pixels in the inspection area are defined. The frequency distribution showing the relationship between the color measurement value and the number of pixels is obtained by performing color measurement using the color elements of the color system, and then the inspection target is compared with the preset reference frequency distribution and the frequency distribution. A color inspection device characterized by determining the quality of a color of an object is disclosed.

Patent Document 3 captures the reflected light obtained by irradiating the surface of an inspection object with light with a color camera, converts the output signal into chromaticity, and obtains the chromaticity and the reference color for the entire inspection area. The color difference from the degree and its frequency distribution are obtained, and the frequency threshold is high for this frequency distribution and the preset small color difference, and the frequency threshold rapidly decreases to zero as the color difference increases. If the entire frequency distribution is less than or equal to the threshold curve, it is judged that there is no color unevenness in the inspection object, and if there is a part that does not fit in the threshold curve, it is judged that there is color unevenness. It discloses a method for inspecting color unevenness, and discloses a standard chromaticity that is the chromaticity obtained for a standard sample.

In Patent Document 4, the parameter setting device maps the color of each pixel of the target image as the target point and the color of each pixel of the excluded image as the exclusion point in the color space (color histogram). Then, a color range that divides the color space and that maximizes the difference between the number of target points and the number of exclusion points (total frequency value) included in the color range is obtained, and the obtained color range is obtained. Set as a color condition (color parameter) used in substrate inspection. As a result, a technique for automatically generating a color condition, which is one of the parameters for inspection, is disclosed.

Patent Document 5 describes an image pickup device having three spectral sensitivities (S1 (λ), S2 (λ), S3 (λ)) linearly converted equivalent to the CIE XYZ color matching function, and a first image obtained by the image pickup device. Generates a normalized two-dimensional or three-dimensional first chromaticity map distribution of the region of interest of one image, and the normalized two-dimensional or three-dimensional region of interest of the second image acquired by the imaging device. A second chromaticity map distribution is generated, the first chromaticity map distribution is compared with the second chromaticity map distribution, and the first chromaticity map distribution and the second chromaticity map distribution are overlapped. An image color distribution inspection device that detects a region, detects the number of first pixels in the region of interest, detects the number of second pixels in the overlapping region, and calculates the ratio of the number of second pixels to the number of first pixels. It is disclosed.

Further, Patent Document 6 discloses a coloring inspection apparatus that clearly and easily quantifies the texture such as metallic feeling and pearl glitter, and rationalizes the comparative inspection between the inspection object and the reference object. Further, Patent Document 7 discloses a color / texture management system and the like that appropriately match colors / textures by utilizing nonstandard products that do not conform to the standard.

Japanese Unexamined Patent Publication No. 2013-33017 Japanese Unexamined Patent Publication No. 9-2063664 Japanese Unexamined Patent Publication No. 2004-144545 Japanese Unexamined Patent Publication No. 2006-78285 Japanese Unexamined Patent Publication No. 2014-187558 Japanese Unexamined Patent Publication No. 2015-155892 JP-A-2018-115877

Patent Document 1 evaluates surface texture such as uneven gloss, and there is a case where the unevenness cannot be sufficiently evaluated for an object to be inspected having unevenness in color due to lack of distribution of unevenness or the like. Patent Documents 2 to 7 perform evaluation based on color information. In these technologies, standard values such as images and standard samples (standard products) are set in advance and compared with the standard values.

However, in the visual inspection, there are cases where the standard value is not determined, or the color changes for each product and there is no standard product or standard value. Further, the color is recognized by entwining a plurality of elements, and the standard value may have a certain width. However, in the case of a product having many regions close to the threshold value of the standard value, the evaluation result by comparison with the standard value may be a defective product, but it may not be said to be uneven when actually observed. Therefore, the standard product and the method using the standard value may not be available depending on the inspected object.

Further, since it is considered that the color unevenness may be caused by the manufacturing conditions of the product and the like, an appropriate evaluation method of the color unevenness is also useful as an examination index for optimizing the manufacturing conditions. However, in the visual evaluation, the color sensitivity may vary depending on the evaluator, physical condition, environment, and other conditions. Depending on the conditions of visual evaluation and the tendency of color, there are colors in which unevenness is conspicuous and colors in which unevenness is inconspicuous. In this way, what appears to be remarkable unevenness in visual evaluation is actually due to a slight difference in product condition, etc., and although it is difficult to distinguish the difference, large unevenness actually occurs, and the result of visual evaluation is the manufacturing process. If you give feedback to, the condition setting may not be enabled. Under these circumstances, there is a demand for an evaluation of color unevenness that is easy to quantify.

Under such circumstances, it is an object of the present invention to provide an apparatus and method for evaluating color unevenness of an inspected object, which can be used for various inspected objects and is easy to recognize and quantify.

As a result of diligent research to solve the above problems, the present inventor has found that the following invention meets the above object, and has arrived at the present invention. That is, the present invention relates to the following invention.

<1> A light source, a half mirror that reflects at least a part of the light emitted from the light source to the surface to be inspected and irradiates the surface to be inspected.
It has an imaging unit that captures an image of the inspection range by capturing light reflected and / or scattered from the inspection range of the surface to be inspected and transmitted through the half mirror.
A detection means for detecting a color component for each pixel of an image captured by the imaging unit, and
A color system conversion means for converting the color component into a color system component for each pixel, and
A reference value setting means for obtaining a reference value in the inspection range from a spatial distribution using at least two characteristic values of the color system component, and
A color unevenness inspection apparatus having an Nx calculation means for calculating a norm Nx in the spatial distribution from the reference value to the pixel for each pixel of an image in the inspection range.
<2> The color unevenness inspection apparatus according to <1>, wherein the color system is an XYZ color system and / or a Lab color system.
<3> The color unevenness inspection apparatus according to <1> or <2>, wherein the reference value is based on any one selected from the group consisting of a mode value, an average value, and a median value.
<4> The color unevenness inspection apparatus according to any one of <1> to <3>, which has a distribution map display means for displaying the distribution of Nx as a distribution map of the inspection range.
<5> The color unevenness inspection device according to <4>, wherein the distribution map display means displays the binarized distribution map by setting a threshold value in the Nx.
<6> The color unevenness inspection apparatus according to any one of <1> to <5>, wherein the object to be inspected is a color-treated stainless steel.
<7> The surface to be inspected is irradiated with a half mirror that reflects at least a part of the light emitted from the light source to the surface to be inspected, and is reflected and / or scattered from the inspection range of the surface to be inspected. A color component detection step of detecting the color component of each pixel of the image captured by the imaging unit that captures an image of the inspection range of the surface to be inspected by imaging the light transmitted through the half mirror, and a color component detection step.
A color system conversion step of converting the color component into a color system component for each pixel, and
A reference value setting step of obtaining a reference value in the inspection range from a spatial distribution using at least two characteristic values of the color system component, and
A color unevenness inspection method comprising a Nx calculation step of calculating a norm Nx from the reference value in the spatial distribution to the pixel for each pixel.

According to the present invention, it can be used for evaluation of color unevenness of various objects to be inspected, and color unevenness can be easily recognized and quantified.

It is an overall schematic diagram for demonstrating the 1st Embodiment of the inspection apparatus which concerns on this invention. It is a flow chart for demonstrating the flow of the inspection method which concerns on this invention. It is an observation image of the stainless steel plate imaged by the optical system which concerns on Example of this invention. It is a figure for demonstrating an example of the norm calculation method concerning the Example of this invention. In the embodiment of the present invention, it is a figure which shows the position of each pixel in the xy space with the object to be inspected as a stimulus value. It is an image which evaluated the color unevenness of the stainless steel colored steel sheet by the inspection apparatus which concerns on Example of this invention. It is another image which evaluated the color unevenness of the stainless steel colored steel sheet by the inspection apparatus which concerns on Example of this invention. It is an observation image of the stainless steel colored steel sheet imaged by the optical system which concerns on a comparative example.

Hereinafter, embodiments of the present invention will be described in detail, but the description of the constituent elements described below is an example (representative example) of the embodiments of the present invention, and the present invention is described below unless the gist thereof is changed. It is not limited to the contents of. When the expression "-" is used in the present specification, it is used as an expression including numerical values before and after the expression.

[Inspection device of the present invention]
The inspection device of the present invention includes a light source, a half mirror that reflects at least a part of the light emitted from the light source to the surface to be inspected and irradiates the surface to be inspected, and an inspection range of the surface to be inspected. It has an imaging unit that captures an image of the inspection range by capturing light that is reflected and / or scattered from the image and transmitted through the half mirror, and detects a color component for each pixel of the image captured by the imaging unit. A reference value in the inspection range is obtained from a detection means, a color system conversion means for converting the color component into a color system component for each pixel, and a spatial distribution using at least two characteristic values of the color system component. It is a color unevenness inspection apparatus having a reference value setting means and an Nx calculation means for calculating a norm Nx in the spatial distribution from the reference value to the pixel for each pixel of the image in the inspection range.
The inspection device of the present invention can be easily adopted for evaluating color unevenness of various objects to be inspected. Further, the inspection apparatus of the present invention can easily recognize and quantify color unevenness.

[Inspection method of the present invention]
In the inspection method of the present invention, the surface to be inspected is irradiated with a half mirror that reflects at least a part of the light emitted from the light source to the surface to be inspected, and the surface to be inspected is reflected and / or reflected from the inspection range of the surface to be inspected. Alternatively, the color component detection step of detecting the color component of each pixel of the image captured by the imaging unit that captures the image of the inspection range of the surface to be inspected by capturing the scattered light transmitted through the half mirror, and the above. A color system conversion step of converting the color component into color system information for each pixel, and a reference value setting step of obtaining a reference value in the inspection range from a spatial distribution using at least two characteristic values of the color system component. This is a color unevenness inspection method including a Nx calculation step of calculating a norm Nx from the reference value in the spatial distribution to the pixel for each pixel.
The inspection method of the present invention can be easily adopted for evaluating color unevenness of various objects to be inspected. In addition, the inspection method of the present invention can easily recognize and quantify color unevenness.
The inspection method of the present invention can be carried out using the inspection apparatus of the present invention, and the configurations corresponding to each other can be carried out with reference to the following.

The present inventors have examined a method for evaluating the color unevenness of an object to be evaluated, which is a new evaluation target for which evaluation criteria have not been established, without providing a reference sample. Therefore, it was examined to set a reference value from the image actually captured for the inspected object by using the stimulus value obtained by converting the image to be inspected into the color system.

As the color system, a color system such as an XYZ color system and a Lab color system is known. These have characteristic values that serve as evaluation indexes, such as stimulus values. The present inventors have studied plotting the evaluation results for each pixel of the image to be observed in a space in which a plurality of these stimulus values are adopted, and setting a reference value from the results. Furthermore, although the stimulus value has multiple axes and the tendency changes depending on the direction, it was examined to uniformly evaluate these as the distance from the value set as the reference value.

In this way, if the reference value set from the image and the norm from the reference value to each pixel are displayed as a distribution map, it becomes easier to perform binarization and grayscale display, and this distribution map is We found that it was easy to recognize and identify unevenness. The present invention is based on such findings, and relates to a method, an apparatus, and the like capable of evaluating unevenness from the inspection range of an inspected object even when a standard sample is not used.

Further, in evaluating the color, it was found that, depending on the type of the object to be inspected, the image of the surface to be inspected may be significantly affected by the influence of illumination on the surface to be inspected (for example, FIG. 8). reference). When imaging the surface to be inspected of the object to be inspected, the XY stage supporting the inspected object is set as a predetermined position, a certain range is imaged, and the XY stage is divided and imaged so as to move. In some cases. When imaging this certain range, a slight difference in the incident angle of the illumination on the object to be inspected may have a great influence on the reflected light and the like, and the image to be imaged may be affected by the slight difference. Further, when a wide range of color unevenness on the inspection surface is evaluated by arranging a plurality of such images, an abnormal portion that does not correspond to the actual unevenness may easily occur.

For example, some of the chemically oxidized color-developing films of stainless steel obtained by the stainless steel color-developing treatment generate an interference color by utilizing the physical properties and film thickness of the plating film. There is also a demand for a technique for evaluating the state of interference color of such a color-developed stainless steel sheet. Generally, when observing surface conditions such as color, in order to suppress the influence of reflected light with a strong amount of light and observe using scattered light, etc., the scattered light of the light radiated diagonally to the surface to be inspected is mainly used. It is often arranged for observation. However, as a result of studies by the present inventors, a stainless steel sheet having an interference color such as color-treated stainless steel is greatly affected by the incident angle of incident light, and it is difficult to inspect color unevenness.

As a result of investigating an irradiation method for an object to be inspected that is easily affected by such interference colors, a half mirror in which the light of a surface light source diffused by a diffuser is arranged substantially perpendicular to the surface to be inspected is used. By irradiating the surface to be inspected from a substantially vertical direction and observing the image through an imaging unit arranged on the extension of the surface to be inspected and the half mirror, even a chemically oxidized and colored film of stainless steel having an interference color is incident. It was found that color unevenness can be inspected by suppressing the influence of angle dependence and the like. The present invention is based on such findings, and even for an inspected object that may be affected by the angle of incidence, it is possible to suppress the influence and evaluate unevenness from the inspection range of the inspected object. Regarding possible methods and devices.

[First Embodiment]
FIG. 1 shows an outline of the overall configuration of the first embodiment of the inspection device according to the present invention. The inspection device 100 is a device that inspects the color unevenness of the inspected body 21 placed on the support portion 20. The inspection device 100 includes an optical system 10, a support unit 20, and an imaging unit 30. Further, the surface inspection device 100 includes an image detection means 40, an image analysis means 50, and a display unit 60 in order to appropriately display and analyze the image captured by the image pickup unit 30. The images captured by the imaging unit 30, the images detected and analyzed by the image detecting means 40 and the image analysis means 50, and the like are appropriately stored in the storage unit 70. Further, the storage unit 70 is used in which conversion formulas, calculation formulas, etc. for detection and analysis are appropriately stored.

[Optical system 10]
The optical system 10 has a structure that illuminates the surface to be inspected of the object to be inspected 21. The optical system 10 includes a light source 11 and a half mirror 13. Further, a diffuser plate 12 is provided between the light source 11 and the half mirror 13. The light emitted from the light source 11 is partially reflected by the half mirror 13 and illuminates the surface to be inspected of the inspected body 21 supported by the support portion 20. A part of the light reflected from the surface to be inspected passes through the half mirror 13 and is imaged by the imaging unit 30.

[Light source 11]
The light source 11 is a light source that irradiates the object to be inspected 21. Light including visible light can be appropriately adopted as the light source 11 depending on the type of the object to be inspected 21 and the purpose of inspection. As the light source 11, a general white light source, an LED light source, a laser light source, or the like can be used. The light source 11 may be provided with a lens (not shown), and the light to be irradiated may be appropriately adjusted and irradiated. As the lens, a lens that converges the irradiation range, a lens that diffuses the lens, a lens that irradiates as parallel light, or the like can be used. Irradiation may be performed without providing a lens depending on the irradiation range and irradiation intensity. As the light of the light source 11, polarized light such as linearly polarized light and elliptically polarized light can be used as appropriate.

The light source 11 is arranged so that light is incident on the reflecting surface in the half mirror 13 arranged on the support portion 20. Further, in the optical system 10, the light source 11 irradiates the diffuser plate 12 with a luminous flux, and the light passing through the diffuser plate 12 is incident on the half mirror 13.

[Diffusion plate 12]
The optical system 10 has a diffuser plate 12. The light from the light source 11 is preferably diffused by the diffuser plate 12. By using a laser light source or an LED light source as the light source 11, highly directional irradiation can be performed. However, since these light sources may be point light sources and the directivity may be too strong, the intensity of the illumination light is determined by using the diffuser plate 12. Irradiate evenly.

[Half mirror 13]
Then, the light emitted from the light source 11 is diffused by the diffuser plate 12 and transmitted to the half mirror 13 side. The light separation surface of the half mirror 13 is arranged obliquely with respect to the optical axis of the light emitted from the light source 11. Further, this arrangement is also arranged diagonally with respect to the optical axis of the reflected light (scattered light) from the object to be inspected. Therefore, the light emitted from the light source 11 to the half mirror 13 and reflected by the separation surface is applied to the surface to be inspected of the object to be inspected 21. Then, depending on the state of the surface to be inspected, the light reflected from the surface to be inspected and incident on the half mirror 13 is determined according to the angle formed by the optical axis of the light reflected from the object to be inspected and the half mirror 13. The light of the irradiation intensity of the above passes through the half mirror 13 and is detected by the imaging unit 30 via the objective lens 31. For this reason, it is preferable that the optical axis of the reflected light from the object to be inspected and the surface on which the half mirror 13 separates the light form 45 °.

By illuminating the surface to be inspected with the object to be inspected 21 by such an optical system 10, the object to be inspected 21 can be illuminated by illumination from a substantially vertical direction of the surface to be inspected. Therefore, it is possible to perform imaging in which the influence of the incident angle dependence such as oblique incidence is suppressed.

[Support 20]
The support portion 20 supports the inspected body 21. The support portion 20 may be a table on which the inspected body 21 is placed, or may have a grip portion that grips the inspected body 21 so as not to move during the inspection. It is also possible to have an XY stage in order to adjust the inspection range of the inspected body 21. The object 21 to be inspected can be, for example, a target for inspecting various molded products, painted products, coated products, plated products, and the like for color unevenness.

[Imaging unit 30]
The imaging unit 30 images the inspection range of the surface to be inspected of the object 21 to be inspected. The imaging unit 30 appropriately scatters and / or reflects light from the inspected object 21 (hereinafter, referred to as “inspected object 21”, depending on the optical system 10, the positional relationship with the inspected object 21, the purpose of inspection, the type of the inspected object 21, and the like. "Scattered light, etc.") is detected. The range of imaging by the imaging unit 30 can be appropriately adjusted by using the objective lens 31. The intensity and color information of the scattered light or the like captured by the imaging unit 30 is appropriately stored in a storage unit or the like for conversion to a color system or the like. The detection of the scattered light or the like can be detected by the area sensor as two-dimensional planar information so that the position of the scattered light or the like of the object to be inspected can be specified. For example, a CCD camera or the like capable of detecting RGB color information for each pixel can be used. In order to reduce variations in the intensity of scattered light and the detection sensitivity of color information, it is better to detect the scattered light and the like a plurality of times and obtain the intensity and color information of the scattered light and the like as an integral value thereof.

Further, the surface inspection device 100 appropriately displays and analyzes the image captured by the imaging unit 30. The image captured by the image capturing unit 30 is detected by the image detecting means 40 as a color component for each pixel of the image, and is analyzed by the image analysis means 50 based on the color component for each pixel. Further, the analysis result is appropriately displayed on the display unit 60.

[Image detection means 40]
The image detecting means 40 is a detecting means that acquires information about an image imaged and transmitted by an imaging unit 30 or the like, and detects an image of an inspection range of an inspected object as a color component for each pixel. The color component may be one that can detect a plurality of colors as long as it can be converted into a color system. For example, the image detecting unit 40 detects the detection result for each pixel such as RGB as the information of each color component for each pixel. The detected color component information is appropriately stored in the storage unit 70, and can be analyzed by the image analysis means 50 or the like, or can be used as the captured image itself for evaluation by comparing with the analysis result.

[Image analysis means 50]
The image analysis means 50 includes a color system conversion unit that converts a color component into a color system component for each pixel. Further, the image analysis means 50 includes a reference value setting means for obtaining a reference value in the inspection range from a spatial distribution using at least two characteristic values of the color system component. Further, the image analysis means 50 includes an Nx calculation means for calculating the norm Nx in the spatial distribution from the reference value to the pixel for each pixel. By performing such an analysis, it is possible to analyze color unevenness and the like as an image of the inspected object 21 from an image in which information on color components is detected by the image detecting means 40 without using a standard sample or the like. Further, the image analysis means 50 may include means for setting a predetermined threshold value for binarization, or may include a discrimination means for extracting a shape feature amount and the like to perform quality determination, classification, and the like.

[Display unit 60]
The display unit 60 uses an image of the analysis result analyzed by the image analysis means 50, a discrimination result related to the image, an imaging result detected by the image detection means 40, information for selecting and operating these, and the like. It is a monitor that displays as appropriate so that people can easily understand it.

[Color unevenness inspection method]
FIG. 2 is a flowchart of the color unevenness inspection method according to the present invention. The color unevenness inspection method S1 according to the present invention includes an irradiation step S11 for irradiating the surface to be inspected with light, an imaging step S21 for imaging the inspection range of the surface to be inspected, and a color component of the image taken. A color component detection step S31 for detecting each pixel, a color system conversion step S41 for converting a color component for each pixel into a color system, and a reference value setting for setting a reference value based on the value of the color system for each pixel. It is possible to have a step S51 and an Nx calculation step S61 for calculating the difference from the reference value for each pixel, and a display step S71 for displaying the difference from the calculated reference value for each pixel. Hereinafter, a case where the color unevenness inspection method S1 is performed using the color unevenness inspection apparatus 100 will be described as an example.

[Irradiation step S11]
The color unevenness inspection method S1 first performs an irradiation step S11 of irradiating the surface to be inspected with light using the optical system 10. In the optical system 10, the light emitted from the light source 11 is diffused by the diffuser plate 12 and incident on the half mirror 13, and a part of the light is reflected by the separation surface of the half mirror 13 to be perpendicular to the surface to be inspected of the object 21 to be inspected. The surface to be inspected is irradiated from the direction.

[Imaging step S21]
Next, an imaging step S21 for imaging the surface to be inspected of the object to be inspected is performed. The light emitted from the vertical direction through the half mirror 13 is reflected or scattered depending on the condition of the surface to be inspected of the object to be inspected 21 or the like. A part of the light reflected / scattered from the surface to be inspected in the direction of the half mirror 13 passes through the separation surface of the half mirror 13 and is transmitted in the direction of the image pickup unit 30, and the image is observed and imaged by the image pickup unit 30.

[Color component detection step S31]
Next, the color component detection step S31 for detecting the color component for each pixel in the inspection range is performed on the image to be inspected captured in the imaging step S21. As the image to be inspected, the image captured by the imaging unit 30 of the inspection device 100 is used. An inspection range is set for these images to be inspected, and a color component for each pixel is detected for the inspection range. The pixel size and the number of pixels are appropriately set in consideration of the size of the inspection target and the like. For example, the size of each side of each pixel can be about 1 μm to 1 mm. It may be 30 pixels or more, 50 pixels or more, or 100 pixels or more per side. The upper limit of the number of pixels per side may not be particularly set, but an upper limit such as 100,000 pixels or less, 50,000 pixels or less, or 10,000 pixels or less may be set.

[Color system conversion step S41]
Next, a color system conversion step S41 for converting the color component into a color system for each pixel is performed with respect to the information regarding the color component in the color component detection step S31. By converting to a color system, it is possible to perform processing that is easier to correspond to a color perceived by a person than using information as it is detected such as RGB. As a result, it is possible to perform an advanced inspection in which the display of color unevenness is consistent with the judgment of a skilled inspector or the like. As the color system, it is preferable to use an XYZ color system or a Lab color system. These are easy to use because a conversion method from data related to color components of a general-purpose detection method such as RGB has been established, and are also suitable as spatial information for setting the reference value of the present invention. ..

[Reference value setting step S51]
Next, a reference value setting step S51 for setting a reference value in the inspection range is performed based on the color system value for each pixel of the image in the inspection range of the color system conversion step S41. To set this reference value, a characteristic value such as a stimulus value of a color system is processed as a spatial distribution for each pixel to obtain a reference value. For example, based on the result of conversion to the XYZ color system, a reference value is obtained from the spatial distribution using X and Y as the stimulus value of each pixel in the image in the inspection range. The reference value can be based on one selected from the group consisting of the mode value, the average value, and the median value. For these mode values, average values, and median values, the values may be used as they are, or based on the values, a certain range before and after the values may be used as a reference, or the correction values may be used as the correction values. May be added or a coefficient is provided. By using two or more stimulus values for setting the reference value in this spatial distribution, the uneven data processing that reflects the color information becomes effective. Further, the reference value setting step S51 may obtain a three-dimensional spatial distribution using three stimulus values. This reference value has an advantage that it can be obtained from the inspection range of an arbitrary object to be inspected even if there is no standard product or the like.

[Nx calculation step S61]
Next, using the reference value set in the reference value setting step S51, an Nx calculation step S61 for calculating the norm (norm Nx) in the spatial distribution with the reference value for each pixel in the inspection range is performed. This norm calculates the norm Nx from the reference value in the spatial distribution to the pixel for each pixel. For example, when the reference values (Xc, Yc) are set using the stimulus values of X and Y, the norm N1 with the conversion value (X1, Y1) of the color system of a predetermined pixel is expressed by the following equation (X1, Y1). Obtained in 1).
N1 = ((X1-Xc) 2 + (Y1-Yc) 2) 0 · 5 formula (1)
That is, the norm N1 obtained by this equation (1) is the absolute value of the distance between the value of each pixel in the spatial distribution and the reference value. By setting the absolute value of such a distance, the data can be easily quantified without displaying the direction in the spatial distribution. Further, when the unevenness is visualized, it becomes easier to recognize the unevenness than the one that also displays the color distribution.

[Display step S71]
Next, a display step S71 for displaying the norm Nx calculated in the Nx calculation step S61 on an image is performed. By this display step S71, the user can confirm the inspection result by seeing the result displayed on the display unit such as a monitor, judge the quality or condition of the product, verify the manufacturing condition of the product, and the like. .. A gradation display such as a gray scale for the norm Nx, or a binarized display by setting a threshold value can be displayed as a distribution map corresponding to an image in the inspection range. By using such a distribution map, it is possible to obtain a distribution map that makes it easier to identify the presence / absence, range, shape, pattern, etc. of color unevenness than an image obtained by simply capturing the inspection range of the object to be inspected. Further, the distribution map using the norm Nx with the reference value according to the present invention can be easily used as data for quantification and the like.

In the present invention, the reference value for obtaining the norm Nx can be set each time within the inspection range of the inspection target. On the other hand, since the norm Nx threshold value can be an absolute value of the norm Nx, the threshold range can be set in advance. For example, when using the stimulus values of X and Y of the XYZ color system, it is possible to set a threshold value as an abnormal point when the norm Nx exceeds 0.005 or as an abnormal point when the norm Nx exceeds 0.01. .. Similarly, in the case of gradation display, for example, an abnormal point is set as a white pixel, a stepwise display is performed up to a predetermined range, and when the distance is more than a predetermined value, a white pixel is displayed or processed. You can also.

By binarizing the image of the inspection range according to the present invention, it becomes easy to determine the quality and the classification. For example, an outlier pixel identification step of specifying outlier pixels (for example, white pixels) outside the threshold of the binarized distribution map and an outlier pixel group consisting of a plurality of outlier pixels adjacent to each other are specified. The shape feature amount extraction step of extracting the shape feature amount indicating the graphic feature of the specified outlier pixel group is provided, and the shape feature amount is extracted by the shape feature amount extraction step using a mapping for the shape feature amount. It is also possible to judge the color unevenness based on the value of the judgment parameter obtained from the value of the shape feature amount. Histograms, multivariate analysis, and the like can be used for mapping.

Further, these steps and means can be executed by a computer as a program. Further, this program can be appropriately stored in a computer-readable recording medium and used.

Hereinafter, the present invention will be described in more detail with reference to Examples, but the present invention is not limited to the following Examples unless the gist thereof is changed.

[Inspection device (1)]
The following experiment was carried out by the inspection device according to the first embodiment. The main configurations in this experiment are as follows.
[Optical system]
A half mirror was placed in the direction perpendicular to the surface to be inspected of the object to be inspected. The half mirror was arranged so that the angle formed by the separation surface of the half mirror in the vertical direction of the surface to be inspected was 45 degrees. A light source was provided via a diffuser plate in the reflection direction of the separation surface of the half mirror. The optical axis of the light emitted from the light source is incident on the half mirror, reflected by the separation surface of the half mirror, and irradiates the surface to be inspected of the object to be inspected from the vertical direction.
-Light source: "LCM-28 / 28W manufactured by Matsudensha" (light source that integrates a flat LED light source, diffuser plate, and half mirror)
[Inspected body]
A SUS304 steel sheet coated with a stainless chemical color treatment film was used as an inspected object, and the inspection range was imaged while being appropriately moved while being placed on the XY stage.
[Image pickup unit]
The subject to be inspected was imaged with a camera. The camera is supposed to be able to detect the intensity of RGB for each pixel. The camera is arranged so that the surface to be inspected of the inspected object arranged horizontally is imaged from the vertical direction and the scattered light of the irradiated light is detected. The pixels of the image to be captured had a size of about 10 μm × 10 μm per pixel. The captured image was detected and stored as having RGB color information for each pixel.
-Camera: The Imaging Source DFK33UX249

FIG. 3 is an image of an inspected object, which is a chemically oxidized color-developing film of stainless steel, taken by using the above optical system and an imaging unit. This image is a side-by-side display of images taken by 7 mm × 5 mm while moving the XY stage. Color unevenness was evaluated for this image.

[Color unevenness evaluation example (1) XYZ color system]
[Reference value setting]
For the image of FIG. 3, RGB color components for each pixel of the captured image were detected. Next, the RGB color components were converted into an XYZ color system for each pixel. FIG. 4 is a diagram for explaining a calculation example for obtaining the norm N1 from the reference value (Xc, Yc) to the value (X1, Y1) of the XYZ color system of a predetermined pixel using the XYZ color system. Therefore, the norm Nx of each pixel can be obtained by performing this calculation for each pixel. FIG. 5 shows the stimulus value X and the stimulus value Y for each pixel converted into the XYZ color system as a distribution. In the image of FIG. 3, when the values for each pixel were examined in terms of the XYZ color system, X was about 0.24 and Y was about 0.255, which was the mode. This mode value was set as the reference value N0 of this example.

[Calculation of norm Nx]
Next, the norm Nx from the reference value N0 to each pixel was calculated for each pixel. This Nx is an absolute value calculated as a distance from N0.
In FIG. 6, the distribution of Nx is displayed in gray scale step by step, and the norm (real number) from N0, which is the mode value, is converted into an 8-bit value of 256 gradations and displayed. .. FIG. 6B shows a binarized distribution of Nx with a threshold value. When the norm (distance) from the reference value exceeds 0.01, this threshold value is displayed as an abnormal point in white pixels, and when the norm is 0.01 or less, it is displayed in black pixels as a normal point.

As shown in FIG. 3, it may be difficult to evaluate color unevenness due to the influence of shading and the like as the image of the chemically oxidized color-developing film of stainless steel. On the other hand, according to the present invention, by setting a reference value within the inspection range and comparing it with the reference value, it is possible to convert the image into an image in which color unevenness can be easily discriminated without using a standard sample. Since FIG. 6 is displayed brighter as it deviates greatly from the reference point, it is an image in which it is easy to identify a part different from the area occupying the majority recognized as the surrounding reference. Since FIG. 6B is binarized, it is an image that makes it easier to more selectively recognize a place having a large norm with the reference region.

[Comparison example]
FIG. 8 is an image taken by irradiating a flat LED light source from an oblique direction (about 45 degrees) with respect to the surface to be inspected without using a half mirror. When the XY stage is set to a predetermined position, the image imaged has a large luminance unevenness due to an incident angle or the like, and the luminance unevenness of each of the divided images is large. Further, when such images are arranged, the joints of the divided images are remarkable. Therefore, it is difficult to effectively evaluate color unevenness.

[Color unevenness evaluation example (2) L * a * b * color system]
[Reference value setting]
For the image of FIG. 3, RGB color components for each pixel of the captured image were detected. Next, the RGB color components were converted into an L * a * b * color system for each pixel. In the L * a * b * color system, L * was about 66.9, a * was about 0.5, and b * was about -31.2, which was the mode. This mode value was set as the reference value N0'of this embodiment.

[Calculation of norm Nx']
Next, for each pixel, the norm Nx'from the reference value N0'to each pixel was calculated. This Nx'is an absolute value calculated as the distance from N0'.
In FIG. 7A, the distribution of Nx'is displayed in gray scale step by step, and the norm (real number) from N0', which is the mode, is converted into an 8-bit value of 256 gradations. Is displayed. FIG. 7B shows a binarized distribution of Nx'with a threshold value. When the norm (distance) from the reference value exceeds 0.01, this threshold value is displayed as an abnormal point in white pixels, and when the norm is 3.2 or less, it is displayed as a normal point in black pixels.

Similar to the example using the XYZ color system, the standard sample using the L * a * b * color system is also obtained by setting a reference value within the inspection range and comparing it with the reference value. It is possible to convert to an image in which color unevenness can be easily discriminated without using. Since FIG. 7A is displayed brighter as it deviates greatly from the reference point, it is an image in which it is easy to identify a part different from the area occupying a large number of people recognized as the surrounding reference. Since FIG. 7B is binarized, it is an image that makes it easier to more selectively recognize a place having a large norm with the reference region.

INDUSTRIAL APPLICABILITY The present invention can be used for evaluating color unevenness of molded products, painted products, coated products, plated products, etc., and is industrially useful.

10 Optical system 11 Light source 12 Diffusing plate 13 Half mirror 100 Inspection device 20 Stage 21 Inspected object 30 Imaging unit 31 Objective lens 40 Image detection means 50 Image analysis means 60 Display unit 70 Storage unit

Claims (7)

A light source, a half mirror that reflects at least a part of the light emitted from the light source to the surface to be inspected and irradiates the surface to be inspected.
It has an imaging unit that captures an image of the inspection range by capturing light reflected and / or scattered from the inspection range of the surface to be inspected and transmitted through the half mirror.
A detection means for detecting a color component for each pixel of an image captured by the imaging unit, and
A color system conversion means for converting the color component into a color system component for each pixel, and
A reference value setting means for obtaining a reference value in the inspection range from a spatial distribution using at least two characteristic values of the color system component, and
A color unevenness inspection apparatus having an Nx calculation means for calculating a norm Nx in the spatial distribution from the reference value to the pixel for each pixel of an image in the inspection range. The color unevenness inspection apparatus according to claim 1, wherein the color system is an XYZ color system and / or a Lab color system. The color unevenness inspection apparatus according to claim 1 or 2, wherein the reference value is based on any one selected from the group consisting of a mode value, an average value, and a median value. The color unevenness inspection apparatus according to any one of claims 1 to 3, further comprising a distribution map display means for displaying the distribution of Nx as a distribution map of the inspection range. The color unevenness inspection device according to claim 4, wherein the distribution map display means displays the binarized distribution map by setting a threshold value in the Nx. The color unevenness inspection apparatus according to any one of claims 1 to 5, wherein the object to be inspected is stainless steel that has been color-developed. The surface to be inspected is irradiated with a half mirror that reflects at least a part of the light emitted from the light source to the surface to be inspected, and the half mirror is reflected and / or scattered from the inspection range of the surface to be inspected. A color component detection step of detecting a color component for each pixel of the image captured by an imaging unit that captures an image of the inspection range of the surface to be inspected by imaging the transmitted light, and a color component detection step.
A color system conversion step of converting the color component into a color system component for each pixel, and
A reference value setting step of obtaining a reference value in the inspection range from a spatial distribution using at least two characteristic values of the color system component, and
A color unevenness inspection method comprising a Nx calculation step of calculating a norm Nx from the reference value in the spatial distribution to the pixel for each pixel.

Images