JP5640812B2 - Paint color evaluation method - Google Patents

Description

The present invention also relates to the coating color valuation method relates to the particular metallic paint, valuation methods of paint colors such as pearl paint.

  Metallic paint, pearl paint, and the like have different colors of appearance (sense when viewed) depending on the direction of light, light intensity (illuminance), viewing direction (angle), and the like. Such a paint color is called anisotropic paint color. Techniques for quantitatively evaluating such anisotropic paint colors that differ depending on the appearance of people have been studied.

  For example, in the evaluation of metallic paint color, a color difference is measured from a plurality of angles by a colorimeter, and a gray zone angle that is not classified into either an approximate region close to the reference color or a non-approximate region is obtained for each angle. There is a technique for evaluating by further obtaining the degree of color change at this gray zone angle (Patent Document 1).

  In addition, for example, there is a technique in which illumination is applied to a painted object from a plurality of directions at different angles, the reflection is received, and evaluation is performed using a color difference formula based on different brightness coefficients in the highlight direction and the shade direction (Patent Document 2).

JP 2009-264924 A JP 2006-292578 A

  However, in any of the conventional methods described above, when a certain color is evaluated, evaluation is performed by obtaining a color difference from a reference plate for evaluating the color. For this reason, every time the color changes, it is necessary to re-determine the range of the color difference obtained by the color difference meter with respect to the visual evaluation as the evaluation standard. For this reason, if the number of colors is large, a lot of labor is required for preliminary verification and preparation for evaluation.

  Therefore, the object of the present invention is to systematically and from the measured values obtained with a color difference meter without depending on the paint color from the measured values obtained with a color difference meter for anisotropic paint colors such as metallic paint and pearl paint. It is to provide a paint color evaluation method capable of quantitatively evaluating a paint color.

In order to achieve the above object, the paint color evaluation method of the present invention classifies the paint color into one of absorption, reflection, and interference according to its color development form, and determines the quality of the paint color for each classification. Create a reference tolerance that will serve as a reference. And the tolerance | permissible_range for determining the quality of the coating color of a specific to-be-evaluated coating material is determined using the reference | standard tolerance | permissible_range of the classification | category to which the coating color of to-be-evaluated coating material belongs. The paint color of the object to be evaluated is evaluated based on this tolerance. Here, the reference allowable range represents a curve indicating the correlation between lightness and lightness difference in a plurality of paint colors whose visual determination results are at the boundary between a non-defective product and a defective product in one classification for each classification, or the curve concerned. It is a formula.

  According to the paint color evaluation method of the present invention, the paint color is classified for each of the coloring forms of absorption, reflection, and interference, and a reference allowable range serving as a reference for evaluation is determined for each classification. Then, from this reference allowable range, an allowable range for each coating color in the color forming form of the classification is determined. And according to the tolerance | permissible_range, it is supposed that the coating color of the to-be-evaluated target object is evaluated. This allows systematic and quantitative tolerances based on the same standards without depending on the specific paint color, even if anisotropic color such as metallic paint or pearl paint has the same color development form. Will be able to decide.

It is a flowchart which shows the procedure of the preparation stage of the coating color evaluation method. It is a flowchart which shows the procedure of the evaluation step of a paint color evaluation method. It is explanatory drawing for demonstrating a measurement angle. It is a graph which shows a part of value of L ^* , a ^* , and b ^* which measured several colors about several coating materials about the coating color classified into each of solid coating, metallic coating, and pearl coating. For metallic coating, (a) is a graph plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) measured at an angle of 15 degrees of “good / bad boundary product”, (b) is a graph of (a) It is a graph which shows the curve obtained by regression calculation. For metallic paint, (a) is a graph plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) measured at an angle of 25 degrees of “good / bad boundary product”, (b) is a graph of (a) It is a graph which shows the curve obtained by regression calculation. It is a graph plotting values of lightness (L ^* ) and lightness difference (ΔL ^* ) measured at angles of 45 degrees, 75 degrees, and 110 degrees of (a) “good / bad boundary product” for metallic coating, (b) These are the graphs which show the curve obtained by the regression calculation of the graph of (a). For pearl coating, (a) is a graph plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) measured at an angle of 15 degrees of “good / bad boundary product”, (b) is a graph of (a) It is a graph which shows the curve obtained by regression calculation. For pearl coating, (a) is a graph plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) measured at an angle of 25 degrees of “good / bad boundary product”, (b) is a graph of (a) It is a graph which shows the curve obtained by regression calculation. It is a graph plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) measured at angles of 45 degrees, 75 degrees, and 110 degrees of (a) “good / bad boundary product” for pearl coating, (b) These are the graphs which show the curve obtained by the regression calculation of the graph of (a). It is the graph which put together the curve of the correlation of each angle in metallic painting. It is the graph which put together the curve of the correlation of each angle in pearl painting. It is the graph which put together the curve of the correlation of each angle in solid painting. It is sectional drawing of the part in which components are attached with respect to the vehicle body. It is sectional drawing of the other part to which components are attached with respect to the vehicle body. It is sectional drawing of the further another part in which components are attached with respect to the vehicle body. a ^* and ^{b *} and _S C, is an explanatory diagram for explaining a relationship between _{S H.} It is a graph which shows the relationship between saturation C ^* and saturation difference (DELTA ⁾ C ^* . It is the graph which showed to what extent the tolerance | permissible_range was extended by multiple when the tolerance | permissible_range of each color was set to 1 from the difference in the component shape of an attachment part about a vehicle type or a site | part. It is a graph which shows the result of having measured the spectral reflectance repeatedly, and trying the classification | category of the coating-film surface which includes a luster material and particle-like reflection is seen. It is a graph which shows the result of having tried the classification | category of the pearl coating color measured by inserting a polarizing film in the light-receiving part of a measuring device in FIG. It is a block diagram for demonstrating a paint color evaluation apparatus. It is the schematic which shows an example of a color filter. It is the schematic which shows an example of the supporting member for performing the variable operation | movement of a camera. It is explanatory drawing for demonstrating the component of the light which reaches | attains a camera in the state without a polarizing filter, when image | photographing a coating surface using a coating color evaluation apparatus. It is explanatory drawing for demonstrating the component of the light which reaches | attains a camera in a state with a polarizing filter, when image | photographing a coating surface using a coating color evaluation apparatus. It is a drawing substitute photograph which shows the image image | photographed without the polarizing filter, the image image | photographed with the polarizing filter, and the image only of the reflected light and interference light derived from the luster material obtained by a subtraction process. It is a graph which shows the detected wavelength and its area ratio about each of the diffuse reflected light from a base surface, the reflected light by a reflecting material, and the interference light by an interference material. Along with the bright material component images obtained in actual metallic coating and pearl coating, the average brightness of the entire image, the enlarged image of representative points of scattered light, the brightness in the enlarged image, the saturation of the entire image, and the particle feeling It is a drawing substitute photograph of the image shown on the display. It is a flowchart which shows the evaluation procedure by the coating color evaluation apparatus. It is a figure which shows the example which installed the camera with respect to multiple angles, respectively. It is a block diagram which shows the example which has arrange | positioned the illuminometer which measures the illumination intensity of a measurement environment. It is a block diagram which shows the example which arrange | positioned the camera which measures the illumination intensity of a measurement environment.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted. Moreover, the size and ratio of each member in the drawings are exaggerated for convenience of explanation, and are different from the actual size and ratio.

[Embodiment 1]
The first embodiment is a paint color evaluation method using systematic and quantitative evaluation criteria that can be applied to anisotropic paint colors such as metallic paint and pearl paint even if the colors are different. In addition, since solid coating can perform reliable quality determination (evaluation) more conventionally, in this embodiment, it demonstrates centering on the evaluation method of metallic coating and pearl coating.

The color difference meter receives the light emitted from the coating film at a specified angle, obtains the spectral intensity for each of a plurality of channels, and converts it into an eye stimulus. In order to express a color, three-dimensional coordinates are often indicated by three values. In the case of painting, the notation by “Lab color space” is generally used, and brightness (L ^* ) and red intensity (a ^* ), Often indicated by yellow intensity (b ^* ). Note red opposite color green, opposite color yellow since it is defined as blue, a ^* is green if negative, b ^* represents the negative if blue intensity.

In order to compare elements other than lightness of colors having different hues, saturation (C ^* ) is defined and calculated by the distance of coordinates represented by a ^* and b ^* . When displayed in saturation (C ^* ), the hue (H ^* ) obtained from the phase angle is defined as auxiliary coordinates.

These are values determined by observing human sensations so that the values and sensations are almost equal. However, even if the distance between two coordinates is obtained from L ^* a ^* b ^*, it often differs from the sense that humans feel as a color difference. Since the sensation of feeling colors as a category broadly differs from the sensation of feeling uncomfortable with the difference between two adjacent colors, it is necessary to convert the colorimetric values displayed in the coordinate system into values representing color differences.

  Furthermore, the color stimulus emitted from the paint film of an automobile is not uniform, and there may be an element that feels a color difference in a surface or a three-dimensional object from a bright part to a dark part. For this reason, it has been difficult to replace human evaluation with colorimetric values alone.

  The paint color evaluation method of Embodiment 1 is roughly divided into a preparation stage for creating a standard necessary for color evaluation, and an evaluation stage for actually judging pass / fail of the color of a paint according to the standard obtained in the preparation stage. Become.

  FIG. 1 is a flowchart showing the procedure of the preparation stage.

  This preparation stage is a process for creating a reference for determining pass / fail of the paint color from the result of measuring the paint color by a color difference meter.

  For this purpose, first, the paint colors to be evaluated are classified according to their color forms (S1). In the present embodiment, the type of paint categorizes whether the light emitted from the coating film to be measured is caused by absorption, reflection, or interference. In general, solid paints absorb light, metallic paints reflect light, and pearl paints produce interference and color.

  The solid coating film is composed of a pigment that absorbs light of a certain wavelength, and is recognized as a uniform color. For example, in a color that looks green when viewed from any direction, a pigment that is absorbed outside the wavelength range of light that appears green is mixed in the paint. Therefore, the coloring form of solid paint is absorption.

  In metallic paint, the original color (called base color) is caused by absorption in the same way as solid paint, but light-reflective metal powder (or metal pieces) is contained in the coating, and white light When hit, the amount of reflection changes depending on the viewing angle, and appears to change from white to gray. Therefore, since such a reflection component is added to the original color, the color development form is reflection. The metal powder (or metal piece) contained is, for example, aluminum or titanium.

  The base color of pearl paint is caused by absorption as well as solid paint. In pearl coating, a light-transmitting substance is contained in the coating film. This light-transmitting substance is composed of a multilayer material having different refractive indexes. Since such a multi-layered material is light-transparent, the light transmitted through the material becomes interference light, and the color itself and the glossiness appear to change depending on the viewing direction and the direction of light. Therefore, since such an interference light component is added to the original color, the color development form is interference. The light-transmitting substance is called mica.

  Metallic powder and multi-layered materials that make up metallic paint and pearl paint are collectively called glitter materials. For this reason, the classification based on the color development form can be replaced with the classification based on the presence / absence of the glittering material and its characteristics. For example, paint colors that do not contain glitter materials (solid paint), paint colors that contain reflective glitter materials (metallic paint), and paint colors that contain light-transmissive glitter materials (pearl paint). it can.

  The size of the glitter material is often 20 to 40 μm in the case of paints for automobiles, both metallic and pearl.

  For each of the above classifications, the quality of the plurality of coated objects is determined visually (S2). This pass / fail judgment is performed for a plurality of colors and a plurality of painted objects within the same classification. The pass / fail judgment is performed visually and is classified into “non-defective product” and “defective product” by an inspector (evaluator). In addition, record how the user feels uncomfortable. For example, white, black, red, yellow, blue, green, even garlic, glare, etc., and the evaluator also records the notation.

  For metallic and pearl paints, there are many paints that are difficult to judge. For example, if there are multiple inspectors, all the inspectors make the non-defective product “good”, and conversely all the inspectors make the defective product “defective”. There is a division. There are also cases in which the inspector is unsure of the decision. In this case, a product whose classification is divided by such an inspector or a product that has a problem is defined as a “good / bad boundary product”. This pass / fail product was named in this way because the distribution of brightness differences in the colorimeter values (to be described later in S3) recorded together is located at the border between “good” and “defective”. In other words, as a distribution of color difference meter values (lightness difference), a set of paint colors that gather at the boundary between “good” and “defective” can be referred to as “good / bad boundary products” in the visual determination result. .

Measurement (also referred to as colorimetry) is performed on the evaluated coated object at a plurality of light receiving angles (S3). The measured value of the color difference meter is recorded in correspondence with the visual evaluation result by S2. This measurement is a color measurement in the L ^* a ^* b ^* color system. The color measurement is displayed by the difference (ΔL ^* , Δa ^* , Δb ^* ) with respect to the reference plate of the target coating color. The brightness difference (ΔL ^* ) obtained here is a difference with respect to a plate determined as a reference for measuring the color difference.

  In addition, the color measurement is to make it possible to sufficiently capture the light from the glittering material included in the base colors of the metallic paint and the pearl paint. Specifically, the range of light received by the color difference meter is set to be at least larger than the interval at which the glittering material is dispersed.

From the colorimetric results of the light receiving angle of the evaluated product, a graph with the lightness coordinate (L ^* ) of the paint color on the horizontal axis and the lightness difference (ΔL) of the evaluated paint on the vertical axis is plotted for each color form of the paint color. (S4). When combined with the determination result of S2, it was found that the value of ΔL ^* determined to be “non-defective” increases as the coating color lightness (L ^* ) becomes brighter (non-defective product in FIG. )reference).

The data to be plotted is obtained by measuring a plurality of painted objects for a plurality of colors within the same classification. Therefore, the plotted results are obtained by arranging samples measured in the same color form and at the same light receiving angle, and classified as “non-defective product”, “defective product”, and “defective product” and ΔL ^* magnitude and lightness coordinates (L ^* ) The relationship is summarized in one graph. However, if it is only for determining the reference allowable range, only the graph for the sample classified as “good / bad boundary product” may be used.

The order of S2-4 may be different. For example, lightness (L value) and lightness difference (ΔL ^* ) are obtained for all painted objects from a plurality of angles, plotted on a graph, and then points other than “good / bad boundary products” are deleted from the graph. May be.

From the result of this plot (S4 above), the correlation between the lightness difference (ΔL ^* ) and the lightness (L ^* value) at the boundary between “non-defective product” and “defective product” (ie, “good product”) (on the graph) And a formula representing the curve) (S5). The correlation may be obtained, for example, by regression calculation for “good / bad boundary products”. This is one of the features of the present embodiment, and since a certain correlation is recognized on the graph, it is expressed as a mathematical expression. In addition, the regression calculation uses existing statistical calculation software to calculate the value of brightness (L value) and brightness difference (ΔL ^* ) at each point on the boundary between “good” and “defective” (“good / bad”). To do. Therefore, from S3 to S5, it is possible to collectively perform the process until the correlation equation is obtained using a computer. For example, brightness values measured for a plurality of painted objects of a plurality of coating colors having a visual evaluation result of “good / bad boundary product” are input to a computer, and a difference in brightness from the brightness of the reference plate is calculated. Thereafter, the computer can be caused to perform a regression calculation from the value of the brightness and the brightness difference to obtain a correlation formula.

Then, the lightness difference (ΔL ^* ) of the obtained curve indicating the correlation is set to an allowable range (referred to as a reference allowable range) as a reference for performing non-defective product determination for each angle in the classification (S6). That is, within the same classification, the absolute value of the lightness difference (ΔL ^* ) of the curve showing the correlation obtained even if the colors are different is a criterion for the quality determination.

Then, using the obtained reference allowable range, an allowable range of brightness difference for performing non-defective product determination is determined for each color (S7). This is based on the correlation curve of the classification to which the color belongs (which may be a correlation formula; the same applies hereinafter), and the brightness indicated by the correlation curve at the value of the lightness (L ^* ) for each angle of the reference plate of that color. The absolute value of the difference (ΔL ^* ) is set as the allowable range.

  This allowable range is a range in which a color difference cannot be discriminated at all if it is a lightness difference within the allowable range by visual observation, as in the examples described later. Therefore, if the acceptable range is used as a numerical standard, it is possible to determine the quality of metallic coating or pearl coating, which has been difficult to determine up to now only by measuring the brightness using a color difference meter.

  This is the preparation stage. This preparatory stage is a systematic and quantitative standard that is independent of color in the classification for each of the coloring forms of absorption, reflection, and interference (ie, solid, metallic, pearl). An acceptable range can be obtained. From the obtained reference allowable range, the allowable range for each individual color can be determined.

  After determining the permissible range as described above, the evaluation of the coated object is performed as an evaluation stage.

  FIG. 2 is a flowchart showing a procedure of an evaluation stage for evaluating the paint color of a painted object.

  First, it is classified whether the paint color of the object to be evaluated is absorption, reflection, or interference (that is, solid paint, metallic paint, pearl paint) (S11).

Subsequently, the lightness (L ^* ) of the object to be evaluated is measured at each light receiving angle with a color difference meter (S12). The lightness (L ^* ) of the reference plate is measured in advance.

Then, the brightness of the reference plate and the measured brightness and brightness difference (ΔL ^* ) are calculated (S13). Thereafter, it is determined whether the value (absolute value) of the obtained lightness difference (ΔL ^* ) is within the permissible range for that color (the permissible range for each light receiving angle obtained in S7) (S14).

  As a result, it is possible to determine whether or not the paint color is good with the same determination standard for each color regardless of the color. In particular, regarding metallic coating and pearl coating, which have been difficult to determine so far, many colors can be determined regardless of the colors in them. Therefore, it is possible to determine the coating of various colors only by obtaining the permissible range once as a preparation stage using the existing painted color. In particular, the same standard can be used for determining colors other than the paint color used in the preparation stage, for example, a new color that has never existed before.

In addition, in the conventional technique, ΔL ^* is used as an example to determine only by lightness coordinates, but it can also be expressed as a comprehensive determination formula including evaluation elements for hue and saturation.

  For example, in the printing industry, it has been said that CIE ΔE2000 can be applied to the color difference determination. In printed products, the color is uniform and the color is often reproduced by absorbing three colors of ink. However, in automobile paints and the like, metallic paints and pearl paints emit reflected light and interference light, so there are many examples that do not match the allowable range determined by CIEΔE2000. However, this time, it was found that the stimulation to humans by reflected light and interference light hardly perceived a color difference with respect to the color reproduced by absorption, and a comprehensive judgment formula that can evaluate this was derived.

  Here, as is well known, the equation (color difference equation) of CIEΔE2000 is defined as the following equation (1), and it is a correction that reflects the results of research on how humans feel color differences for each hue, brightness, and saturation. The term is set. In the formula, ΔE2000 is denoted as ΔE00.

In the formula, C is saturation, H is hue, ΔL is lightness difference, ΔC is saturation difference, ΔH is hue difference, K _L , K _C , and K _H are parametric coefficients, S _L , S _C, and S _H is the lightness difference, a chroma difference, and heavy valence factor respective each of the hue difference, RT is rotation function. A value added with a bar above each symbol represents an average value of a color difference pair of two colors. Therefore, here, it is an average value of the values of the reference plate and the object to be evaluated. The parametric coefficients K _L , K _C , and K _H are values that are appropriately set according to the test conditions, but are all set to 1 under the standard conditions. Hue angle and θ are degrees.

In this embodiment, this comprehensive judgment formula is created by naming “ΔEn09”. Overall judgment formula, by using the formula Derutai00, the value of S _L of therein are those already with the use of the value of the expression of the correlation relationship described. S _C , S _H , and RT are common to ΔE00 for solid coating, metallic coating, and pearl coating. As a result, it is possible to perform the evaluation associated with the stimulus of the human appearance even in the metallic paint or the pearl paint while continuing the relation with ΔE00.

  Here, the purpose of paint color management in automobiles is to prevent the customer from feeling uncomfortable when all parts are assembled into a vehicle. For this reason, it is important to determine whether or not each angle is good, but it is necessary to quantitatively evaluate not only the allowable range that determines good and bad for each angle, but also the appearance of the paint color of the entire automobile body. For example, since a body, a bumper, and other color parts are painted with a paint specified in accordance with the paint color specification, the generated color difference is restricted within the range of the paint specification. More specifically, the color difference is caused by a painted condition or a paint lot difference.

  Therefore, there is a similarity in the light emitted from the paint film between the paint specifications, that is, the paint color specified, and as already explained, each of the body, bumper, and other color parts, If solid paint, metallic paint, and pearl paint are classified according to absorption, reflection, and interference, and a relational expression that can be quantitatively determined when viewed as a whole is determined, the measurement result of the color that becomes the standard of paint color Based on the above, it is possible to derive an acceptable range that almost matches the visual evaluation.

  Furthermore, it is possible to evaluate the difference in color between components attached to the automobile body. For example, between the various parts mounted on the car body, visual illusion is caused by the difference in surface and curved surface, as well as the size of the gaps between parts, division by malls, and discontinuity by character lines. Sometimes you get up and the colors look different. However, by quantitatively managing the brightness difference and / or ΔEn09 value as in the present embodiment, the color management for the shape can be quantitatively evaluated even when the color looks different due to the illusion phenomenon. Is possible.

[Example of Embodiment 1]
Here, an example will be described in which a threshold value corresponding to visual evaluation is obtained from a measurement result actually using a color difference meter.

For the color measurement of the coating color, a multi-angle spectrocolorimeter MA-68II manufactured by x-rite was used. FIG. 3 is an explanatory diagram for explaining the measurement angle. Here, five angles devised for the paint color whose brightness and color change depending on the viewing direction are shown. As shown in the drawing, the light source was arranged at a position of elevation angle θ = 45 degrees from the painted surface. The light source was installed 1000 mm away from the illumination center 0 (zero) on the painted surface. The light source is white light. The measurement angles were 0 degree in the regular reflection direction and 15 degrees, 25 degrees, 45 degrees, 75 degrees, and 110 degrees as the angles toward the light source direction, and the lightness L ^* was measured at each angle. The measurement position is a position of 500 mm from the illumination center 0 (zero) on the painted surface in the direction indicated by the arrow at each angle.

FIG. 4 is a chart showing a part of values of L ^* , a ^* , and b ^* in which a plurality of colors are measured for a plurality of painted objects for the paint colors classified into solid paint, metallic paint, and pearl paint, respectively. It is. Here, about 100 samples (total of about 300 points) were measured for each classification. Note that the measured values shown in FIG. 4 are examples of measured values of L ^* , a ^* , and b ^* for the sake of understanding of the present example. In all of the values measured as the present example, Absent.

FIGS. 5 to 7 are graphs plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) of “good / bad boundary products” (vertical brightness difference (ΔL ^* ), horizontal brightness (L ^* ) ^. )) And a curve obtained by the regression calculation. In each figure (a) is the lightness (L ^*) and lightness difference ([Delta] L ^*) graph plotting the values of, in each figure (b) is a graph showing a curve obtained by regression calculation. FIG. 5 is a graph at an angle of 15 degrees (only this angle is plotted with “good (◯)” and “defective (Δ)” for reference). FIG. 6 is a graph of an angle of 25 degrees. FIG. 7 is a graph of angles 45 degrees, 75 degrees, and 110 degrees.

Next, FIGS. 8 to 10 are graphs plotting values of brightness (L ^* ) and brightness difference (ΔL ^* ) of “good / bad boundary products” (vertical brightness difference (ΔL ^* ), horizontal brightness for pearl coating. (L ^* )) and a curve obtained by the regression calculation. In each figure (a) is the lightness (L ^*) and lightness difference ([Delta] L ^*) graph plotting the values of, in each figure (b) is a graph showing a curve obtained by regression calculation. FIG. 8 is a graph with an angle of 15 degrees. FIG. 9 is a graph with an angle of 25 degrees. FIG. 10 is a graph of angles 45 degrees, 75 degrees, and 110 degrees.

  FIG. 11 is a graph summarizing correlation curves of angles in metallic coating. FIG. 12 is a graph summarizing correlation curves of angles in pearl coating. For further reference, FIG. 13 is a graph summarizing correlation curves obtained by the same method for solid coating.

  From the graphs of FIGS. 11 and 12, for metallic coating (reflection) and pearl coating (interference), as the value of brightness of the coating color (horizontal axis) increases, the allowable width (vertical axis) of the color difference increases. I understood it. On the other hand, such a relationship was not recognized in solid coating (absorption).

When measuring the brightness of the reference plate in silver metallic coating, the reference plate L ^{* of} 15 degrees is 130, the reference plate L ^* of 25 degrees is 95, the reference plate L ^* of 45 degrees is 65, and the reference plate L of 65 degrees. ^* Was 44, and the reference plate L ^{* of} 110 degrees was 38.

When this value is applied to the graph of metallic coating shown in FIG. 11, at 15 degrees, the value of the lightness difference (ΔL ^* ) on the vertical axis is 8.0 at the position of the horizontal axis (L ^* ) 130. . Therefore, the allowable range of the lightness difference (ΔL ^* ) is ± 8.0. Similarly, ± 4.8 at 25 degrees, ± 3.0 at 45 degrees, ± 2.8 at 75 degrees, and ± 1.8 at 110 degrees.

Further, the allowable range was obtained in the same manner for other colors. The gray metallic paint, 15 degrees of the reference plate ^{L *} is 83,25 ° of the reference plate ^{L *} is 66,45 ° of the reference plate ^{L *} is 39,75 ° of the reference plate ^{L *} is 21,110 ° of the reference plate L ^* Was 14. When these values are applied to the graph of metallic coating shown in FIG. 11, the allowable range of the lightness difference (ΔL ^* ) is ± 6.3 at 15 degrees, ± 4.0 at 25 degrees, and ± 3.3 at 45 degrees. It is ± 1.5 at 0 and 75 degrees and ± 1.0 at 110 degrees.

Similarly, an allowable range was obtained for pearl coating. In pearl white coating, the 15 ° reference plate L ^* is 106, the 25 ° reference plate L ^* is 94, the 45 ° reference plate L ^* is 85, the 75 ° reference plate L ^* is 84, and the 110 ° reference plate. L ^* was 183. When these values are applied to the graph of pearl coating shown in FIG. 12, the allowable range of the lightness difference (ΔL ^* ) is ± 5.8 at 15 degrees, ± 2.5 at 25 degrees, and ± 2. It is ± 2.0 at 0 and 75 degrees and ± 2.0 at 110 degrees.

Similarly, in pearl gray paint, the 15 ° reference plate L ^* is 39, the 25 ° reference plate L ^* is 30, the 45 ° reference plate L ^* is 17, the 75 ° reference plate L ^* is 7, 110 °. The reference plate L ^* was 5. When these values are applied to the graph of pearl coating shown in FIG. 12, the allowable range of brightness difference (ΔL ^* ) is ± 4.5 at 15 degrees, ± 2.8 at 25 degrees, and ± 1. It is ± 0.9 at 1,75 degrees and ± 0.5 at 110 degrees.

As a result of visual evaluation of each coating color, no difference in color from the reference plate was recognized even though all the above-described coating colors had different brightness differences (ΔL ^* ) within the allowable range. Therefore, it was found that the above-described allowable range is in good agreement with how the color is visually perceived.

As a result of the evaluation based on the above-described allowable range, when items determined as non-defective products are collected, the position is shifted in a certain direction ((−L) direction or (+ L) direction) from 0, which is the median value of the allowable range. The distribution may be biased. In such a case, the absolute value where the center of the allowable range is 0 is not used, but the median value is shifted in the (−ΔL) direction or the (+ ΔL) direction, and a threshold value in each direction is provided. Also good.

Specifically, for example, when there is a deviation between the lightness of the reference plate and the median value in the set of the lightness values of the coated objects due to the relationship between surface matching and painting reproduction, the median value is obtained from the distribution, and as a result, 15 ° and so on ΔL ^* +1.5 in ΔL ^* +2.0,75 degrees in ΔL ^* -4.0,45 degrees. These values are values for correcting the median value.

If the threshold value of ΔL ^* is calculated by reflecting this median value, it becomes −12 to +4 at 15 degrees. The range is 16, and ± 8.0 in the above allowable range where the median is 0, and the width of the allowable range is the same. Similarly, +45 to +4.8 at 45 degrees and +1.3 to +4.3 at 75 degrees.

  There are various reasons why the deviation from the reference plate must be specified, but one of them is that the light receiving part of the measuring device is displaced due to surface alignment or gaps at the mounting position of the component, and for each angle. It may occur when an accurate value cannot be measured.

  14-16 is sectional drawing of the part in which components are attached with respect to the vehicle body. Examples of the parts include various parts such as bumpers, mat guards, front and rear fenders, and lamps.

  FIG. 14 shows a state in which the vehicle body and the parts are aligned so as to form a continuous curve in design, and the normals of the parts and the vehicle body are directed in the same direction with respect to the visual evaluator. In this state, the normal direction of the boundary region of the butt portion is the same. However, since it is a continuous curve as a whole, there may be a deviation from the case where a flat surface such as a reference plate is measured.

  FIG. 15 shows a case where there is no manufacturing error for both the vehicle body and the parts, but the vehicle body and the parts are misaligned in the boundary region of the butt portion at the time of attachment. In this case, the normal direction of the boundary line region is different even in the allowable range of attachment error. Then, a phenomenon is observed in which the component side gets on the vehicle body, the normal direction changes, and the color of the component looks different from that in FIG. For this reason, a deviation may occur from the case of measuring the reference plate.

  FIG. 16 shows a case where a manufacturing error has occurred in a part of the part. In this case, there is no deviation at the time of attachment, but the normal direction of the boundary line region is also different. In such a case, a phenomenon in which the color of the component appears different from that in FIG. In this case as well, there may be a deviation from the measurement of the reference plate.

  If the median value deviates due to the state of assembly of parts to the vehicle body, etc., if that value is corrected and threshold values that are upper and lower limits are set in the allowable range, quantitative evaluation can be made even more reliably. It can be performed.

  Next, an example of evaluation using ΔEn09 will be described.

  Here, as described above, the multi-angle spectrocolorimeter MA-68II manufactured by x-rite is used for the brightness measurement. This apparatus can measure the color difference received at five angles as shown in FIG.

  The following formula (2) is a calculation formula used for the multi-angle spectrocolorimeter MA-68II manufactured by x-rite in order to calculate the value of ΔEn09. Here, it is shown as a determinant.

Here, the subscripts 15, 25, 45, 75, and 110 indicate angles. The subscript N indicates the painted object to be evaluated. Lt, at, bt, Ht, and Ct are values for correcting the above-described median value. In the above-described example, only the brightness has been described. Similarly, for a ^* and b ^* , values for correcting the median are obtained, and Ht and Ct are converted from values for correcting the median of a ^* and b ^*. That's fine. If there is no need to correct the median value, these values of Lt, at, bt, Ht, and Ct may be omitted. L ^* , a ^* , and b ^* are values obtained by measuring the color of the paint.

Also in the calculation formula in this embodiment, as already described, S _C , S _H , and RT may be values common to solid coating, metallic coating, and pearl coating. The values of K _L , K _C and K _H are 1.

Then, the value of S _L, as the following equation, so that the expression for the curve of correlation fall.

S _L of a solid coating are shown in the following equation (3). Each angle may be a common value.

S _LN = 1 + 0.015 (L ^* −50) ² / {20+ (L ^* −50) ² } (3)
S _L in metallic paint are shown in the following (4) to (8) for each angle. Ln is a natural logarithm (the same applies hereinafter).
S _LN15 = 1.2499Ln (L ^* ₁₅ ) -0.036 where 8 <L ^* ₁₅ <125 (4)
S _LN25 = 0.4524Ln (L ^* ₂₅ ) +1.0638 where 8 <L ^* ₂₅ <105 (5)
S _LN45 = 0.2712Ln (L ^* ₄₅ ) +0.5517 where 8 <L ^* ₄₅ <90 (6)
S _LN75 = 0.4379Ln (L ^* ₇₅ ) +0.029 where 8 <L ^* ₇₅ <90 (7)
S _LN110 = 0.4229Ln (L ^* ₁₁₀ ) +0.1342 where 8 <L ^* ₁₁₀ <90 (8)
SL in pearl coating is as shown in the following formulas (9) to (13) for each angle.
S _LN15 = 3.3395Ln (L ^* ₁₅ ) −8.3341 where 50 <L ^* ₁₅ <150 (9)
S _LN25 = 2.1574Ln (L ^* ₂₅ ) -4.948 where 35 <L ^* ₂₅ <110 (10)
S _LN45 = 1.83Ln (L ^* ₄₅ ) -39581 where 20 <L ^* ₄₅ <70 (111)
S _LN75 = 1.4613Ln (L ^* ₇₅ ) -2.6403 where 12 <L ^* ₇₅ <50 (12)
S _LN110 = 0.9213Ln (L ^* ₁₁₀ ) +-1.1547 where 8 <L ^* ₁₁₀ <42 (13)
By using the above-described equations (3) to (13), a color difference value by ΔEn09 can be obtained. As already described, the allowable range is an allowable range in color difference if ΔEn09 is set by applying the upper and lower thresholds of the allowable range for each classification to the equation.

  In CIEΔE2000, calculation is performed based on the concept using ΔC and ΔH, but the allowable range can also be determined using Δa and Δb. That is, the allowable range is applied as the maximum value / minimum value of Δa and Δb.

Figure 17 is an explanatory diagram for explaining the relationship between a and b and _S C, _{S H.} As shown in FIG. 17 (a), when the two-dimensional orthogonal coordinates a and _b, S C, S _H is represented as a vector indicating the elliptical range for proportional straight line a and b. S _C a and b directly proportional linear direction vectors, S _H is a vector of a direction orthogonal to the proportional straight line.

Using this relationship, as shown in FIG. 17B, the values of a and b in the ellipse defined by Sa (S _C ) and Sb (S _C ) are within the allowable range of Δa and Δb. Here, Sa (S _C), as illustrated, is a vector comprising an a-axis direction of one side of the equilateral triangle and the longer side of the S _C direction of the vector. Meanwhile, Sb (S _C) is a vector of the b-axis direction of one side of the equilateral triangle and the longer side of the S _C direction of the vector. Therefore, the length in the a-axis direction and the length in the b-axis direction of Sa (S _C ) and Sb (S _C ) are within the allowable range of Δa and Δb. The value of this _{S C} is the value of _{S C} used in equation (1). And when the allowable range of this, if seeking the value of S _C from the upper limit value and the lower limit value of the allowable range of the brightness difference described above (or median upper limit threshold correcting) the value the so can be used as the allowable range of S _H.

Further, as shown in FIG. 17C, the values of a and b in an ellipse defined by Sa (S _H ) and Sb (S _H ) can be set within the allowable range of Δa and Δb. Here, Sa (S _H), as illustrated, is a vector comprising an a-axis direction of one side of the equilateral triangle to the long side a vector of S _H direction. Meanwhile, Sb (S _H) is a vector of the b-axis direction of one side of the equilateral triangle to the long side a vector of S _H direction. Therefore, the length in the a-axis direction and the length in the b-axis direction of Sa (S _H ) and Sb (S _H ) are within the allowable range of Δa and Δb. The value of this _{S H} also a value of _{S H} to be used in (1). And when the allowable range of this, if seeking the value of S _H from the upper limit value and the lower limit value of the allowable range of the brightness difference described above (or median upper limit threshold correcting) the value the so can be used as the allowable range of S _H.

Furthermore, an allowable range based on a C value representing saturation can be determined. FIG. 18 is a graph showing the relationship between saturation C ^* and saturation difference ΔC ^* . From this graph, it can be seen that the relationship between the saturation C ^* and the saturation difference ΔC ^* is proportional. Further, there is a relationship of (C ^* ) 2 = (a ^* ) 2+ (b ^* ) 2. By using these relationships, the allowable range based on the lightness difference ΔL ^* is converted into a saturation difference ΔC ^* (absolute value) from the relational expression used when obtaining the above-described CIE ΔE2000 and ΔEn09. The tolerance range by value can be determined. Similarly, it is possible to convert to an allowable range by ΔH.

  As in the above examples, in addition to the color difference systemization due to lightness, saturation, and hue, the difference between metallic and pearl in multi-angle colorimetry is also systematized to maintain the same index, You can now compare how you feel the difference and color difference.

  As a result of further visual evaluation for each vehicle type and each part using the formula and value obtained in this example, it was found that the threshold value could be further expanded (depending on the object, it may be narrowed). There were some that I could not get).

FIG. 19 is a graph showing a multiple of how much the allowable range can be expanded when the allowable range of each color is set to 1 due to the difference in the part shape of the mounting part for the vehicle type and the part.
In the figure, an example with an angle of 75 degrees is shown. The narrowest was 0.8 times, and the widest was 1.8 times. Similarly, at other angles, the allowable range width may be expanded or narrowed in correspondence with the visual evaluation.

  In FIG. 19, Z34, L53E, S51, V36, CV36, and HV36 are symbols for indicating the vehicle type. BODY (and Body) is the vehicle body, BMPR (and Bmpr) is the bumper, FDR is the fender, FR (and Fr) is the front, and RR is the rear.

  By accumulating the results of the visual evaluation in this way, it is possible to strictly define a threshold value caused by a difference in vehicle shape due to a difference in vehicle type, a combination of parts, a difference in assembly error, or the like. This is also the result that the paint color can be handled quantitatively and systematically by the present embodiment and examples. In other words, by defining the allowable range for each color form of the coating, it is possible to quantitatively understand how the result of the visual evaluation is different. Therefore, the difference can be shown as a multiple.

  According to Embodiment 1 and Examples described above, the following effects are obtained.

In the first embodiment, first, light emitted from a paint color is classified into one of absorption, reflection, and interference according to its color development form. Then, for each classification, a reference allowable range is created in which measurement results output as values (L ^* , a ^* , b ^* , and derived values) converted from optical intensity are associated with visual evaluation. And the tolerance | permissible_range for determining the quality of the coating color of a specific to-be-evaluated coating material is determined using the reference | standard tolerance | permissible_range of the classification | category to which the coating color of to-be-evaluated coating material belongs. It was decided to evaluate the coating color of the object to be evaluated based on this tolerance. As a result, even in anisotropic coating colors such as metallic coating and pearl coating, systematic and quantitative tolerances can be determined on the same basis for each color form. Therefore, when evaluating anisotropic paint colors such as metallic paint and pearl paint, the paint colors are classified according to the color forms, and the permissible standard that is determined in advance for each color form is used. You should evaluate.

  This is particularly effective when new colors are added that have never been seen before. Conventionally, when a new color is added, it is necessary to re-determine the range of the color difference obtained with the color difference meter for visual evaluation of the new color compared to the reference plate. However, by using the present embodiment and examples, it is possible to evaluate a new color simply by obtaining an allowable range for determining the specific quality of a new color using the already obtained reference allowable range. become able to.

  In addition, the reference allowable range is a curve indicating the correlation between brightness and brightness difference due to a plurality of painted objects of a plurality of colors that are unacceptable in one category for each category, or an expression representing the curve. This is a systematic clarification of the relationship between brightness and brightness difference for paint colors that have been unknown until now. By using this correlation as a criterion for quality determination, it is possible to obtain an evaluation standard for anisotropic coating colors such as metallic coating and pearl coating, which have been difficult to determine.

  In addition, when determining the permissible range for each color, since the standard permissible range is determined for each classification of the color development form, if the brightness of the reference plate for the paint color of the object to be evaluated is known, the value is also included. In addition, the permissible range for individual colors can be easily determined from the standard permissible range. This means that, in the evaluation of the paint color, the paint color is classified according to the color form, and the allowable range of the color is determined using the reference allowable range of the color form to which the color belongs. For this reason, in particular, even for a color such as a new color for which the acceptable range for determining pass / fail has not been specifically determined, it is possible to easily determine and evaluate the acceptable range as long as there is a reference plate for that color. it can.

  In addition, if the median of the lightness difference between the paint colors of the painted products that have been evaluated as non-defective products is deviated from 0 (zero), which is the median absolute value, after the evaluation of the paint colors, A more appropriate upper and lower limit threshold value can be obtained simply by correcting the median without changing the value.

  Furthermore, the reference allowable range obtained for each color form can be corrected so as to be a more accurate allowable range by accumulating the results of visual inspection.

  Further, the classification according to the color form can be easily classified as to which color form the paint color belongs to by classifying according to the presence / absence of the glitter material contained in the coating film and the characteristics of the glitter material.

  In Embodiment 1 and Examples described above, it is assumed that a reference plate prepared in advance is used. However, the present embodiment and examples can be implemented without using a reference plate prepared in advance. For example, in the case of evaluating the paint color of an automobile body, the painted part of the automobile body that has been painted is considered to be equivalent to the reference plate, the brightness is measured, and this is the brightness of the reference plate. And And as already explained, it is only necessary to determine and evaluate the permissible range by the color used as the reference plate from the standard permissible range in the color form to which the coating belongs. The color used as such a reference plate is also a reference color.

  Furthermore, this embodiment is applicable also to the color matching at the time of repainting a part of automobile body, for example. Specifically, for example, a plurality of paints (color-matched paints) that are considered to be close to the paint color of the automobile body to be repainted in advance are prepared, and a sample for each color in which the color is painted on a steel sheet is created. On the other hand, the brightness of the paint on the automobile body is measured, and this is used as the brightness of the reference plate (reference color). Then, using the reference allowable range of the coloring form to which the paint color of the automobile body belongs, the allowable range for the color is determined from the measured brightness of the automobile body. The tolerance is then used to evaluate each sample to determine the color to use. As a result, it is possible to determine a paint color that hardly changes in appearance from the original color of the car body.

  In this embodiment, if there is a measuring instrument (colorimeter) that can measure at least the brightness, the paint color can be evaluated.

  In the above-described embodiment, the paint color evaluation method is roughly divided into a preparation stage and an evaluation stage. However, this is for explanation, and such a stage does not need to be performed in the implementation. For example, the reference evaluation range for each classification is determined in advance by steps S1 to S6 in the preparation stage. Then, at the stage where the paint or color to be evaluated has come out, the individual color evaluation range is determined (S7). After that, it is performed to evaluate the paint color of the individual paint (S11 to 14).

  Various other modifications are possible.

[Embodiment 2]
In the first embodiment described above, it is assumed that the paint color specification as an evaluation target is known. Therefore, since the glittering material used can be specified from the coating color specification, it can be easily understood whether the coloring form of the coating color is classified as absorption, reflection, or interference.

  However, classification may not be easy depending on the paint color. That is, when the paint color specification is unknown, it is a state in which it is necessary to classify whether it is a solid paint, a metallic paint, or a pearl paint. However, classification may be difficult by appearance alone.

  In the second embodiment, a color difference meter is used to classify whether the coating to be evaluated is metallic coating or pearl coating.

  FIG. 20 is a graph showing the results of repeated spectral reflectance measurements and attempts to classify the coating surface that includes the glittering material and shows particulate reflections, for each angle and wavelength in metallic coating and pearl coating. The reflectance is shown. In the figure, the figure in the (a) column is a graph of metallic paint (metallic silver), and each graph is 15 to 110 degrees. (B) The figure of a row | line is a graph of pearl coating (white pearl), and is a graph of 15 to 110 degree | times, respectively. In each figure, the vertical axis represents the reflectance, and the horizontal axis represents the measurement position on the measurement object (generally at intervals of 1 mm (not strictly 1 mm, but may be approximately 2 mm)). A white plate was used as the reference plate, and the reflectance with respect to the white plate was measured using a multi-angle spectrocolorimeter MA-68II manufactured by x-rite. Using a white light source, the reflected light was dispersed every 20 nm in a wavelength range of 400 to 700 nm to obtain reflectance for each wavelength. In each graph, the reflectance for each wavelength is shown together (one line corresponds to one wavelength).

  In the metallic coating and the pearl coating, as shown in FIG. 20, if the measurement is repeatedly performed while shifting the measurement surface of the measurement object, a difference in measurement values coming from finely different distributions of particles can be obtained. In metallic coating and pearl coating, since the glitter material is dispersed in the coating film, when fixed point measurement is used, it may happen that the reflected light or interference light component cannot be measured from the glitter material. By measuring this while relatively moving the light irradiation point on the painted surface and the position of the painted surface, reflected light and interference light from the glittering material can always be captured.

  From the diagram in the row (a) of FIG. 20, the metallic coating has little variation in reflectance due to the difference in wavelength. On the other hand, it can be seen that the reflectance of pearl paint varies greatly depending on the wavelength. In particular, there is a significant difference when the angle is 15 degrees or 25 degrees close to the regular reflection direction (see FIG. 3).

  Therefore, in order to classify metallic paint and pearl paint, it is determined that the paint is pearl paint from the reflectances of a plurality of wavelengths when the reflectance varies for each wavelength, particularly when the reflectance varies by 15 degrees or 25 degrees. can do. On the other hand, if the reflectance for each wavelength is almost the same, the coating is metallic. The degree of variation is less than ± 1% in the metallic coating as shown in FIG. Note that such a range of variation may be determined by comparing with a variation in reflectance of a paint color whose classification is known in advance. For example, the reflectance of a paint color known to be metallic paint (reflection) is compared with the paint color of unknown classification. As a result, if there is a large variation in the paint color whose classification is unknown, it is determined that the paint is pearl paint. On the other hand, if the variations are similar, the paint color whose classification is unknown is determined to be metallic paint.

  Such a difference in reflectance is caused by a difference in the glitter material included in the paint color. The metallic paint includes a reflective glitter material as a glitter material. When the light hitting the bright material is reflected on the surface of the bright material, the reflection direction does not change depending on the wavelength. For this reason, the metallic coating has almost the same reflectivity at any wavelength in any of the above. On the other hand, the pearl paint contains a light-transmitting glitter material. Refraction occurs in the light transmitted through the light-transmitting glitter material. The refractive index varies depending on the wavelength. For this reason, the direction of the light emitted from the glittering material is different for each wavelength. For this reason, the reflectance is different for each wavelength.

  FIG. 21 is a graph showing the result of attempting to classify the pearl paint colors measured in FIG. 20 by inserting a polarizing film in the light receiving part of the measuring instrument. (A) The figure column shows the vertically-changing film, (b) the figure in the column shows the transverse polarizing film, and (c) the figure in the column shows no polarizing film. Note that the vertical is a direction along the regular reflection direction from the light source shown in FIG. 3 (the direction from the right to the left in the drawing), and the horizontal is a direction orthogonal to this. The measurement itself is the same as in FIG.

  As shown in FIG. 21, by measuring through the polarizing film, the difference in the measurement position and the fluctuation due to the irregular reflection by the coating surface (the fluctuation in the horizontal axis direction) are suppressed, and the difference in reflectance due to the difference in wavelength is good. I will understand. In order to suppress such irregular reflection on the painted surface, measurement is performed with a sheet-like film containing a lens for ensuring that light is aligned in a certain direction, a light-shielding film with an open pinhole, etc. The same effect can be obtained (referred to as a light regulating member).

  As described above, in the second embodiment, light having a plurality of wavelengths is applied to the coating object to be evaluated, and the color development forms are classified based on the reflectance. For this reason, it is possible to classify paint colors whose color development form is unknown. After the classification, it is possible to perform pass / fail judgment for the paint color whose color development form is unknown by performing the paint color evaluation of the coated object according to the first embodiment described above.

  The reflectance was measured while relatively moving the position of the painted surface and the measurement point. Thereby, the light (wavelength) from the bright material in a coating film can be caught reliably.

[Embodiment 3]
This embodiment 3 is for systematically and quantitatively evaluating the paint color evenly with respect to anisotropic paint colors such as metallic paint and pearl paint, even if the kind and composition of the glittering material and pigment constituting the paint color are different. This is a paint color evaluation device.

  FIG. 22 is a block diagram for explaining a paint color evaluation apparatus. This paint color evaluation apparatus includes a light source (illuminating means) 11, a variable color filter 12, a first polarizing filter 13, a reflecting mirror 14, a second polarizing filter 15, a lens 16, and a camera (photographing means) 17. Further, a computer (image processing means) 18 for processing an image taken by the camera 17 is connected.

  The light emitted from the light source 11 passes through the variable color filter 12 and the first polarizing filter 13 and is guided toward the painting surface 100 by the reflecting mirror 14 (the light is reflected and applied to the painting surface 100). Light from the painted surface 100 is photographed by the camera 17 through the second polarizing filter 15 and the lens 16.

  The light source 11 is a white light source and illuminates the painted surface. The light source 11 is preferably a white light source having no spectral distribution bias, and for example, a halogen lamp is used. In addition, a white LED lamp or the like may be used as long as the spectral distribution is not biased. The position of the light source 11 is adjustable.

  The variable color filter 12 is arranged so that filters of various colors block the light from the light source 11. The position of each color filter can be changed, and by changing the position of the color filter, the color (that is, the wavelength) of the light emitted from the light source 11 can be changed so that the paint surface 100 can be irradiated. Yes. FIG. 23 is a schematic diagram illustrating an example of the variable color filter 12. As shown in FIG. 23A, the variable color filter 12 has color filter films 122 of different colors attached to a plurality of holes provided along the outer periphery of the disk 121. The disk 121 is supported by a column 123. A disk 121 is rotatably attached to the end of the column 123. FIG.23 (b) is the figure seen from the arrow b direction of Fig.23 (a). As shown in the figure, the position of the color filter film 122 can be changed by rotating the disk 121, and only the wavelength of the color filter film 122 can be transmitted from the light (white light) from the light source 11. It should be noted that at least one of the color filter films 122 is not covered with a filling film or a color filter film when the light from the light source 11 is transmitted as it is. That is, white light can be irradiated as it is. Alternatively, the variable color filter 12 itself or the disk 121 is removed so that white light can be irradiated as it is.

  Note that a band pass filter that allows only a specific wavelength to pass may be used instead of the variable color filter 12.

  The first polarizing filter 13 and the second polarizing filter 15 are polarizing filters, and their operation will be described later. The installation position of the first polarizing filter 13 may be placed anywhere in the light path from the light source 11 to the coating surface 100. The second polarizing filter 15 may be placed anywhere in the light path from the paint surface 100 to the camera 17.

  The reflecting mirror 14 can reflect the light from the light source 11 and change its direction. For this reason, the direction of the reflecting mirror 14 is freely changeable.

  The lens 16 adjusts the focus so that the camera 17 is focused on the painting surface 100, and is provided with an aperture and a shutter (not shown) for adjusting the exposure time.

  The camera 17 includes an image sensor (color image sensor) and captures at least a still image. The camera 17 is preferably a digital camera in order to process the captured image by the computer 18. This camera 17 is supported so that the angle at which the painted surface 100 is photographed can be changed by moving the position of the dotted line shown in FIG.

  FIG. 24 is a schematic view showing an example of a support member for performing the variable operation of the camera 17. As shown in the figure, the support member includes a guide 171 having a semicircular shape (an arc is a part of a perfect circle). The camera 17 moves along the guide 171 and is attached so that the camera 17 always faces the center of the semicircle. The guide 171 is engraved with an angle scale (not shown). Thus, the photographing angle with respect to the painted surface 100 can be easily changed by simply moving the camera 17 along the guide 171. Also, the shooting angle can be determined by the angle scale. Further, the height of the camera 17 from the painted surface 100 can be adjusted as appropriate.

  The installation angle of the camera 17 is the same as that of the first embodiment. For example, when the light irradiation angle from the light source 11 is an elevation angle θ = 45 degrees with respect to the paint surface, the regular reflection direction is 0 degree, and the light source direction The angle to the angle is 15 degrees, 25 degrees, 45 degrees, 75 degrees, 110 degrees, and the like. Of course, shooting at a desired angle is also possible.

  The computer 18 is an image processing means and processes an image photographed by the camera 17. For this purpose, image processing software is introduced into the computer 18, and the paint color of the object to be evaluated is evaluated according to the procedure described later. Images taken with the camera 17 can be sent directly to a computer. The computer 18 is connected to a display (not shown). The image from the camera 17 may not be sent directly to the computer 18, but may be temporarily stored in a server and sent to the computer 18 in another place such as a management room via a network.

  The operation of this paint color evaluation apparatus will be described below.

  As described in the first embodiment, the coating film can be classified into absorption forms for solid coating, reflection for metallic coating, and interference for pearl coating, as described in Embodiment 1. This is determined by the characteristics of the glittering material contained in the paint film. The solid paint does not contain glitter materials. In metallic coating and pearl coating, a glittering material is dispersed in the coating film. For this reason, in the case of a device that measures the lightness of a specific point, such as a color difference meter, since a single light receiving element is usually used, all incoming light must be regarded as the same light. Therefore, when the size of the measurement range on the painted surface 100 is smaller than the density at which the glitter material is dispersed, reflected light or interference light from the glitter material may or may not be captured. On the other hand, if the measurement range is wide, only the color in which the base color, reflected light and interference light are mixed is captured, and only the light from the glittering material cannot be captured separately.

  Therefore, in the third embodiment, a two-dimensional image of the painted surface 100 is photographed by the camera 17, and light from the paint (base surface), reflected light from the bright material, or interference is captured from the image photographed with the resolution of the camera 17. It was decided to separate light by light.

  FIGS. 25 and 26 are explanatory views for explaining components of light reaching the camera 17 when the painted surface 100 is photographed using the paint color evaluation apparatus. FIG. FIG. 26 is an explanatory diagram when photographing is performed without the two polarizing filters 13 and 15, and FIG. 26 is an explanatory diagram when photographing is performed with the first and second polarizing filters 13 and 15.

  In the figure, a bright material that reflects light (referred to as a reflective material) and a bright material that transmits light and refracts and emits interference light (referred to as an interference material) are depicted in the same coating film. This is for illustrative purposes. Some coatings contain both glitter materials, but in many cases, the metallic coating is only a reflective material and the pearl coating is only an interference material.

  Then, as shown in FIG. 25, when the first and second polarizing filters 13 and 15 are photographed, the diffuse reflection light from the base surface (that is, the color due to absorption), the reflection light from the reflection material, and the interference material Enters the camera 17 in a state where the interference light from is mixed. For this reason, since the photographed image (described later) is mixed with those light components, the light from the base surface cannot be separated from the reflected light and interference light from the glittering material.

  However, the image taken by the camera 17 includes light from the base surface, reflected light from the bright material, and interference light. Therefore, if the resolution of the image sensor of the camera 17 is at least high enough to separate the light from the reflecting material and the interference material, the light from each can be separated later. However, without a polarizing filter, no matter how high the resolution is, it is impossible to tell which pixel of each pixel in the image sensor contains light from the base surface, reflected light from a bright material, or interference light. .

  On the other hand, as shown in FIG. 26, when the first and second polarizing filters 13 and 15 are photographed, the reflected light from the paint surface reaches the camera 17, but the reflected light from the reflective material and the interference material The interference light does not reach the camera 17.

  Here, the changing direction (vibration direction) of illumination light from the light source is various directions and does not have a fixed polarization direction. On the other hand, the reflected light on the base surface includes part of the reflected light from the painted surface, but most is diffused light. This is because most of the light component in the direction in which the polarization direction of the reflected light on the painted surface is parallel to the painted surface is reflected as regular reflected light, so the illumination direction of the illumination light from the light source and the camera This is because it hardly comes in because of the position. As shown in the drawing, the illumination direction from the light source is 45 degrees with respect to the painted surface, and the camera position is perpendicular to the painted surface. On the other hand, the polarization direction of light entering the coating film is a direction perpendicular to the painted surface. Then, a part of the light in the polarization direction perpendicular to the coating surface entering the coating film is reflected or refracted by the glittering material in the coating film and comes out of the coating surface. For this reason, by controlling the direction of the polarizing filter, it is basically possible to block only the light reflected by the polarization direction light perpendicular to the painted surface. Therefore, the image when the polarizing filter shown in FIG. 26 is photographed is only diffuse reflection light from the paint surface, and reflection light and interference light derived from the glittering material are not photographed (does not reach the camera 17). Can be.

  Here, in principle, one polarizing filter may be provided in front of the camera 17, but in order to cope with the inclination of the painted surface to be measured and the change in the polarization direction of light due to the bright material, the third embodiment. Then, two of the first polarizing filter 13 and the second polarizing filter 15 are used. The angles of the relative polarization directions of the first polarizing filter 13 and the second polarizing filter 15 are adjusted so that the light derived from the bright material becomes the darkest position (the first polarizing filter 13 and the second polarizing filter 13). This is not the position where the polarizing filter 15 is perpendicular and completely dark).

  From these, by subtracting the image photographed with the first and second polarizing filters 13 and 15 from the image photographed without the first and second polarizing filters 13 and 15 (image processing), The image of only the reflected light and the interference light can be obtained.

  FIG. 27 is a drawing-substituting photograph showing an image photographed without a polarizing filter, an image photographed with a polarizing filter, and an image of only reflected light and interference light derived from a bright material obtained by subtraction processing.

  These images are taken in the relationship between the installation positions (respective angles) of the light source 11 and the camera 17 such that each light irradiation from the light source 11 is at an elevation angle of 45 degrees with respect to the painting surface 100 and the camera 17 is perpendicular to the painting surface 100. In the direction, the incident angle is 45 degrees with respect to the illumination light from the light source. From the relationship between the installation positions of the light source 11 and the camera 17, the specular reflection component from the coating film surface does not enter the camera 17 with or without the polarizing filter, and the base surface inside the coating film. Indirect reflection components (diffused light) from the light, reflected light and interference light derived from the bright material will enter. The installation position of the camera 17 is 70 mm from the painted surface 100 (h in FIG. 22), the resolution of the camera 17 (imaging device) is 1024 × 768 pixels, and the device area is 4.8 × 3.6 mm. The shooting range on the painted surface 100 is set to be 9 mm square. The illuminance of the light source 11 is 7200 lx (lux). The image was taken with a lens aperture value of 16. The exposure time (shutter speed) is adjusted when each image is captured so that the light of the glittering material component does not saturate (saturate). Note that saturation may be avoided by adjusting the aperture while fixing the exposure time. Here, not to saturate is to prevent the signal level of the pixel from exceeding the maximum value within the range of the dynamic range of the image sensor. For example, in the case of 8-bit color, the gradation is 0 to 255. Therefore, when the pixel signal value obtained from the image sensor becomes 255 (maximum gradation value), the exposure time or aperture is adjusted so as to be less than that. In particular, in metallic coating, reflected light from a reflective material that is a bright material may appear strongly. In such a case, the exposure time or aperture is adjusted as appropriate so that the pixels are not saturated, and an image is taken with an appropriate exposure amount.

  For this reason, when obtaining values of lightness and saturation used for evaluation, it is necessary to convert signal values (tone values) from pixels so that images are captured with the same exposure amount. If the dynamic range of the camera is sufficiently large, the aperture and exposure time can be made the same regardless of the paint color. If the aperture and exposure time are the same, conversion of the signal value is not necessary.

  As shown in FIG. 27A, an image taken without the first and second polarizing filters 13 and 15 is an image in which light from a glittering material is scattered in an image having a grainy feeling as a whole. It has become. And as already explained, this image is in a state where the diffused light from the base surface, the reflected light from the reflecting material, and the interference light from the interference material are mixed.

  As shown in FIG. 27 (b), the image taken with the first and second polarizing filters 13 and 15 has an overall grainy image, but the glittering material shown in FIG. 27 (a). There is no point of light by.

  Therefore, as shown in FIG. 27C, the image of FIG. 27B is subtracted from the image of FIG. 27A, so that the glitter material shown in FIG. 27A is scattered. An image from which only the light component is extracted (referred to as a bright material component image) can be obtained.

  Image processing is performed by image processing software in the computer 18. The processing is performed by photographing the same position with / without the first and second polarizing filters 13 and 15, and from the gradation value of each pixel of the image without the first and second polarizing filters 13 and 15. The gradation value of each pixel of the image with the second polarizing filters 13 and 15 may be subtracted.

  By this subtraction process, the light of the bright material component can be separated. However, it is not possible to separate whether the glitter material is a reflection material or an interference material.

  Here, since the reflected light component from the reflecting material is total reflection, all the wavelengths of white light are uniformly reflected when white light is applied. On the other hand, interference light is light that has been refracted in various ways through a light-transmitting bright material (interference material). Moreover, since the refractive index varies depending on the wavelength, the received light component includes a wavelength different from the color of the base surface.

  Therefore, the wavelength of the scattered light of the glittering material component extracted as shown in FIG. 27C is analyzed as a whole, that is, by calculating the saturation of the image, Whether the light is from the material or the interference light from the interference material can be separated.

  FIG. 28 is a graph showing detected wavelengths and their ratios (area ratios) for each of diffuse reflection light from the base surface, reflection light by the reflection material, and interference light by the interference material. Figure (a) is a graph showing the wavelength, and (b) is a graph showing the area ratio.

  As shown in the figure, the diffusely reflected light from the base surface has a lightness peak at the wavelength emitted by the pigment on the base surface (the wavelength of the color not absorbed by the pigment). The reflected light from the reflector is similarly affected by the pigment and has a peak at almost the same position as the peak wavelength from the pigment. However, the brightness of the other wavelengths increases uniformly because of the white light as a whole. doing. On the other hand, it can be seen that the interference light has a wavelength peak at a position different from the diffuse reflection light from the base surface and the reflection light from the reflector. Further, from FIG. 28 (b), the proportion of these lights is known. Diffuse reflected light from the base surface occupies the largest area. The interference light and the reflected light from the reflecting material are less than the diffuse reflected light. This is consistent with the metallic and pearl paints, where the color you feel is the base color, and you can see the feeling of glittering in the color (metallic feeling) and the feeling of various color changes (pearly feeling). To do.

  FIG. 29 shows the average brightness of the entire image, the magnified image of representative points of scattered light, the brightness of the magnified image, the saturation of the entire image, together with the glitter component images obtained in actual metallic coating and pearl coating. 4 is a drawing-substituting photograph of an image in which a particle feeling is projected on a display connected to a computer 18. The light source 11, the camera position, the illuminance, and the like are the same as when the image shown in FIG. 27 is taken.

  Here, the average brightness (shown as “average:%” in the drawing) is an average value of brightness values of each pixel of the entire image, but here, it is shown as a relative value with respect to the reference white diffuse reflector. Yes. Here, the exposure time is adjusted so that the signal value (gradation value) when the white diffuse reflector is photographed is the same as the signal value (gradation value) of the brightest point when the subject paint color is photographed. To do. And the relative value of each exposure time was made into the relative value of the brightness. The same applies to the brightness of the enlarged image. Saturation (value of “particle saturation” in the figure) is converted from the R (red), G (green), and B (blue) signal values in the enlarged image into HSV, and the S (color) image is used for saturation. Is a calculated value. The particle feeling (value of “particle feeling: dB” in the figure) is dB = 20 × log (the highest luminance but its luminance (peak value) / total average luminance) for the pixel. This particle feeling is used as an index representing a rough feeling that a person feels when looking at the painted surface 100. The smaller this value, the less the particle feeling (grainy feeling).

  As shown in FIG. 29 (a), in white pearl which is a pearl coating, the average brightness is 64%, the brightness is 388% for the enlarged pixel 1 ("enlarged 1" in the drawing, the same applies hereinafter), the brightness is 876% for the enlarged pixel 2, and the particles The saturation was 13.68 and the particle feeling was 22.7 dB.

  As shown in FIG. 29B, in the dark blue pearl which is a pearl paint, the average brightness is 3%, the brightness of the enlarged pixel 1 is 65%, the brightness of the enlarged pixel 2 is 88%, the particle saturation is 10.3, and the particle feeling is 27. 0.7 dB.

  As shown in FIG. 29 (c), in gray ash silver metallic which is a metallic paint, the average brightness is 21%, the brightness of the enlarged pixel 1 is 313%, the brightness of the enlarged pixel 2 is 214%, the particle saturation is 2.12, and the particle feeling is 27. 0.7 dB.

  As shown in FIG. 29 (d), in the dark gray metallic as a metallic paint, the average brightness is 5%, the brightness of the enlarged pixel 1 is 173%, the brightness of the enlarged pixel 2 is 147%, the particle saturation is 12.17, and the particle feeling is 28. It was 9 dB.

  From these results, the lightness of white pearl or gray ash metallic is higher than that of dark blue pearl or dark gray metallic due to the influence of the brightness of the original base color.

  On the other hand, the value of particle saturation indicating the overall saturation is larger in pearl coating than in metallic coating. This indicates that the pearl paint contains more wavelength components with different base colors than the metallic paint.

  Based on the above, it is possible to photograph the object to be evaluated with or without a polarizing filter and subtract them to obtain the brightness, saturation, etc. from the light image of the glittering material component. For paint colors that have been difficult to evaluate quantitatively, such as paint and pearl paint, it is possible to reliably evaluate them as numerical values such as brightness and saturation, which are indicators of those colors.

  The above is the basic operation of the paint color evaluation apparatus of the present embodiment, but this apparatus can change the shooting angle of the camera 17 as already described. If the painted surface 100 is photographed at various angles using this angle changing function, evaluation for each angle can be performed.

  Further, by using a color filter, it is possible to irradiate only light of a specific wavelength (specific color) and perform evaluation only for the wavelength.

  FIG. 30 is a flowchart showing an evaluation procedure by the above-described coating color evaluation apparatus.

  First, the light source 11 (orientation of the reflecting mirror 14) and the camera 17 are installed with respect to the painted surface 100 so as to obtain an angle to be evaluated (S21). Then, the images with and without the first and second polarizing filters 13 and 15 and with the first and second polarizing filters 13 and 15 are taken and loaded into the computer 18 (S22). Up to this point, the work is manual.

  Thereafter, processing is performed by the computer 18 in accordance with a preprogrammed procedure. The processing of the computer 18 is S23-25 of this flowchart.

  The computer 18 subtracts the images with the first and second polarizing filters 13 and 15 from the acquired images without the first and second polarizing filters 13 and 15 (S23).

  Subsequently, the computer 18 calculates the average brightness of the entire image, the saturation of the entire image, and the particle feeling from the obtained image after subtraction processing (bright material component image) (S24). Note that, in the evaluation of the paint color in the coated object, it is not necessary to calculate the lightness and saturation of each scattered light in order to evaluate the entire paint color. Note that the brightness and saturation of each of the scattered light can be calculated at this stage for reference.

  Thereafter, the computer 18 takes the difference between the overall brightness and saturation obtained in the same manner according to the reference color of the reference plate in advance and the brightness and saturation of the object to be evaluated obtained in S4, and the brightness difference and A saturation difference is obtained, and a pass / fail judgment result is output depending on whether or not these values are within a predetermined threshold range (S25). The reference color of the reference plate is prepared in advance for the paint color to be evaluated, but may also be a color obtained from the reference coating. Further, if it is only necessary to know the values of lightness and saturation of the object to be evaluated without performing the pass / fail judgment, this S25 need not be executed.

  The results of S24 and S25 are stored in a designated area on a predetermined recording medium. Further, it may be displayed on a display or printed out as necessary.

  According to the third embodiment described above, the following effects are obtained.

  The painted surface 100 of the object to be evaluated is photographed by the camera 17 that captures a wider range than the measurement range by the color difference meter, with or without the polarizing filter. Then, only reflected light and / or interference light is extracted from the bright material in the coating film by taking the difference between the images with and without the polarizing filter. Then, the brightness and / or saturation of the image in which the extracted light exists is determined. As a result, the color of the glittering material component can be systematized by the same operation and quantitatively evaluated for the coating color in which the glittering material is dispersed, such as metallic coating or pearl coating.

  Shooting with and without the first and second polarizing filters 13 and 15 first includes light from the base surface and light from the glittering material without the first and second polarizing filters 13 and 15. Take a picture. Next, the first and second polarizing filters 13 and 15 are put in and an image of only the light from the base surface is taken. Then, by taking the difference between these images (subtraction process), an image containing only light from the glittering material was separated. Thereby, an image including only light from the glittering material can be easily obtained. Then, by this subtraction process, when there is no light from the glittering material, it is solid coating, when there is light from the glittering material, it is metallic coating or pearl coating, and when the saturation is large, it can be determined that it is pearl coating. it can. It should be noted that the photographing order with and without the first and second polarizing filters 13 and 15 may be reversed.

  Further, by making it possible to change the positions of the camera 17 and the light source 11, it is possible to measure brightness and saturation at various angles.

  In the third embodiment, the positions of the camera 17 and the light source 11 are changed. Instead, a plurality of cameras and light sources are prepared, and the camera and the light source are set at the positions of the determined shooting angles. You may make it install. FIG. 31 is a diagram illustrating an example in which cameras are installed for a plurality of angles. As shown in the drawing, cameras 172 to 175 are installed for each determined measurement angle with respect to the arc-shaped guide 171. In the figure, the number of cameras is four. However, the number of cameras is not limited to that shown in the figure as long as it is appropriately prepared in accordance with a required angle. The positions of the cameras 171 to 174 may be movable so that they can be finely adjusted.

  In this way, by installing a camera for each of a plurality of measurement angles, the brightness and saturation values at the desired measurement angles can be continuously and quickly reduced without changing the position of the camera or light source. Can be obtained. This is particularly suitable for a mass production line or the like, and can reduce the evaluation (inspection) time of the paint color.

  Similarly, a plurality of light sources having different angles of light emitted from the light sources may be provided. Also in this case, by switching a plurality of illumination lights having different angles and shooting with the camera, it is possible to save work time associated with changing the camera and the light source position.

  Further, by taking a picture through the variable color filter 12, it is possible to quantitatively evaluate characteristics due to various wavelengths. In the present embodiment, the variable color filter 12 is installed in front of the light source 11 to change the wavelength of light for proving the coating circle. Alternatively, the same effect can be obtained by disposing the variable color filter 12 in front of the camera 17 to limit the wavelength of light entering the camera 17. Therefore, the installation position of the variable color filter 12 may be placed anywhere in the light path from the light source 11 to the painting surface 100 or in the light path from the painting surface 100 to the camera 17.

  In the third embodiment, an illuminometer for measuring light in the measurement environment, another camera, and the like can be further arranged.

  FIG. 32 is a block diagram illustrating an example in which an illuminance meter that measures illuminance in a measurement environment is provided. The configuration other than the provision of the illuminance meter is the same as that described in FIG.

  When the illuminance meter 177 is installed, the position is on the painted surface 100 where light from the light source 11 does not directly enter. In addition, when it has to install in the position where the light from the light source 11 enters directly, it is good to turn off a light source and to measure the illumination intensity of an environment.

  FIG. 33 is a block diagram illustrating an example in which a camera for measuring illuminance in a measurement environment is provided. The configuration other than the provision of the camera is the same as that shown in FIG.

  When the camera 178 is installed, it is provided at a position where at least the range of the painted surface taken by the camera 17 can be taken.

  Such an illuminometer 177 and other cameras 178 are referred to as ambient light measuring means. By measuring the light in such a measurement environment, even when the environment for evaluating the coating color changes, it is possible to prevent the evaluation result from being different depending on the environment by correcting it.

  For this, for example, the illuminance of the ambient light is changed, and depending on the camera 17, a white diffuse reflector is photographed. At this time, an illuminometer that measures the illuminance of ambient light is installed in the vicinity of the shooting position of the camera, or another camera is installed at a position where the entire measurement system including the camera 17 can be photographed to measure the illuminance of the ambient light To do. Then, the difference between the illuminance obtained by the illuminometer or another camera and the overall brightness of the image obtained by photographing the white diffuse reflector is taken as a correction value.

  When evaluating the paint color of the object to be evaluated, the lightness and saturation obtained by photographing the painted surface of the object to be evaluated may be corrected with this correction value.

  Further, instead of such correction, the brightness of the environmental illumination may be changed so that the illuminance of the environment obtained by the illuminometer or another camera is always constant.

  Although the embodiment has been described above, the present invention is not limited to this. For example, the first, second, and third embodiments are a paint color evaluation method and a paint color evaluation apparatus that can be individually implemented. For example, the paint color can be evaluated by combining them. For example, the paint colors of the object to be evaluated are classified according to the second embodiment, and the permissible range serving as the reference and the permissible range of individual colors are obtained in advance according to the first embodiment. Thereafter, the lightness and saturation of the paint color of the evaluation object are measured by the paint color evaluation apparatus of the third embodiment. Then, it may be easy to determine whether or not values such as lightness and saturation measured by the paint color evaluation apparatus are within an allowable range obtained in advance. In addition, the scope of the present invention is determined by the matters described in the claims, and is not construed as being limited to the contents exemplified as the embodiment, and various modifications are possible.

11 Light source (illumination means),
12 variable color filters,
13 Polarizing filter,
15 Pre-camera polarizing filter,
16 lenses,
17 Camera (photographing means)
18 computer (image processing means),
100 painted surface,
121 discs,
122 color filter film,
123 struts,
171 Guide.

Claims (7)

Classifying the paint color into one of absorption, reflection, and interference according to its color development form, and creating a reference tolerance that serves as a reference for determining the quality of the paint color for each classification;
Using the reference allowable range of the classification to which the paint color of the evaluation target paint belongs, determining an allowable range for determining the quality of the paint color of the evaluation target object;
A step of evaluating the paint color of the object to be evaluated according to the allowable range;
I have a,
The reference allowable range is a curve indicating the correlation between lightness and lightness difference in a plurality of paint colors whose visual determination results are at the boundary between a non-defective product and a defective product in one of the classifications, or the curve concerned. paint evaluation wherein the expression der Rukoto representative. The allowable range of the paint color of the object to be evaluated is the absolute value of the lightness difference obtained by applying the lightness of the reference color of the paint color to a curve indicating the correlation between the lightness and the lightness difference or an expression representing the curve. The paint color evaluation method according to claim 1 , wherein the paint color evaluation method is a value. When the allowable range of the paint color of the object to be evaluated has a deviation from the median of the distribution of brightness differences obtained by accumulating the evaluation results, the median value of the allowable range is included in the median of the deviation The coating color evaluation method according to claim 2 , wherein the allowable range is corrected so as to match. The paint color evaluation method according to any one of claims 1 to 3 , wherein the classification is the classification based on presence / absence of a glitter material and characteristics of the glitter material. The classification is characterized by irradiating a coating with an unknown classification with light, measuring the reflectance for each wavelength in the light, and classifying the interference when the reflectance varies. Item 5. The coating color evaluation method according to any one of Items 1 to 4 . 6. The paint color evaluation method according to claim 5, wherein the reflectance is measured while relatively moving the position of the paint surface and the measurement point. The paint color evaluation method according to claim 5 or 6, wherein the reflectance is measured through a light regulating member that suppresses irregular reflection on a painted surface.