RU2292037C2 - Method of recording goniometric characteristics of paint-and-varnish film - Google Patents

Abstract

FIELD: investigating or analyzing materials.

SUBSTANCE: method comprises using device that is provided with the unite (3) for forming an image for recording the interaction of light (reflection or transmission) with the surface, light source (2), and place for setting the specimen with paint-and-vanish film (6) to be analyzed. The unit for forming the image, light source, and specimen are arranged so that, in one image, at least one of the characteristics of the surface can be recorded as a function of continuous range of angles between the illumination direction (7), (9) and observation direction (4), (5). The unit for forming the image is made of a camera based on the charge-coupled device.

EFFECT: enhanced reliability.

8 cl, 8 dwg

Description

The invention relates to a method and apparatus for registering visual properties of a surface, for example color, gloss, texture, etc., of colored coating films using an image forming apparatus, a light source, and a sample region for arranging a sample whose surface is to be examined. Visual signs of the surface may depend on the optical geometry, which is determined by the location of the observed object relative to the observer and the light source. This dependence can occur, for example, in coating films containing spectacular dyes. The dependence of visual properties on optical geometry is usually called goniochromatism or “flop behavior”. In optical geometry, the direction of observation is the line between the observer and the observed point of the observed object, and the direction of illumination is the line between the light source and the observed point. The direction of reflection is the line obtained by specular reflection of the direction of illumination relative to a plane perpendicular to the surface of the sample. The flop angle, also referred to as the aspect angle, is the angle between the direction of observation and the direction of reflection. In addition, the luster of spectacular or effectless films depends on the optical geometry.

Spectacular dyes are used in coatings to produce optical effects, for example, metallic or pearlescent. Typically, in a coating film containing metallic dyes, the brightness depends on the optical geometry, while in coatings containing pearlescent dyes, the tint changes with optical geometry. This complicates the characterization of the visual characteristics of such coating films. An additional complication in this regard is the appearance of spots with local strong scattering when observing the coating film at a short distance, due to the presence of dye flakes. While pure color can be characterized by the spectral distribution of reflectance values, coatings containing spectacular dyes require taking into account the angular and spatial dependences.

Until now, the dependence of the brightness and color of metallic and pearlescent coatings on optical geometry has been investigated using a spectrophotometer for a limited number of different geometric measurement configurations. However, this leads to an incomplete picture providing data only for a very limited number of optical geometric configurations.

US Pat. No. 5,550,632 discloses a method and apparatus for evaluating paint films using a digital camera. Only one optical geometry is recorded at a time, and since the camera is focused, only one flop angle. Since the characteristics of the coating layer containing effective dyes depend on the flop angle, this method cannot be used to evaluate such effective coatings in a single recording cycle.

The objective of the invention is a system that allows you to evaluate the surface with a single registration cycle over a continuous range of flop angles.

To solve the problem of the invention, there is provided a device for recording surface characteristics depending on the flop angle, comprising an image forming device for detecting the interaction of light with the surface, a light source and a sample area for receiving a sample, the surface of which is to be studied, characterized in that the coverage range of the forming device image covers a continuous range of directions of observation and the fact that the imaging device, the light source and the region of the sample It is arranged in such a way that in one image at least one of the characteristics of the surface is recorded as a function of the flop angle. Registration can be used for visual assessment and comparison, or, depending on the ability of the imaging device to measure the recorded signals, for measuring or processing data. Particular examples of use are gloss measurement and color matching.

The device or system according to the present invention, in particular, is designed to assess the flop behavior or goniochromatism of the surface, with a coating containing effective dyes.

In order to obtain a useful picture of flop behavior, the range of flop angles defined above should preferably span more than 40 degrees, more preferably more than 50 degrees.

For some types of surfaces, such as coating films containing metallic dyes, the color is darker at large flop angles. For use in this case, the full range of measurements, the light distribution preferably varies in the viewing range of the image forming apparatus, preferably in accordance with an increasing or decreasing function. This function depends on the type of material. The distribution of light may vary by changing the light output of the light source as a function of the angle of illumination. Alternatively, the light distribution may be varied by using appropriate filters.

In a preferred configuration, the light source may be a linear source, for example a TL strip light source, a horizontal slit in the light diffuser, an array of point sources, such as LEDs or glass fibers, etc. Alternatively, the light source may be a point source.

A suitable imaging device in a configuration consistent with the invention is a CCD camera (charge coupled devices). Suitable CCD cameras are, for example, Ricoh ^® RDS 5000, Olympus ^® C-2000Z, Minolta ^® Dimage ^® RD 3000 and Nikon ^® Coolpix ^® 950.

Another group of devices for use in the present invention is constituted by digital video cameras. Thanks to the use of a video camera, the flop angle can change not only as a function of position, but, alternatively or additionally, as a function of time. Using a video camera also allows you to track time-dependent changes in the visual characteristics of the test surface over a period of time, for example, the appearance of the coating film during curing.

When using a digital camera, each registered image consists of a large number of pixels. Each pixel has a red value of R, a green value of G, and a blue value of B. Ideally, the calibrated values of R, G, and B for an absolutely black surface should be 0, and for an ideal absolutely white surface, each of these values should be equal to the specified maximum value. The maximum value is 2 ⁿ -1, where n is the number of bits that specify the pixel. When using an 8-bit pixel depth, the maximum value is 255.

In the study of metal coatings, the brightness can locally exceed the level of white. This must be taken into account, for example, choosing the maximum value of white color less than 2 ⁿ -1.

For accurate color measurement, it is preferable to periodically calibrate the measurements. Using a camera on a CCD, for calibration, for example, you can first individually record a black sample and a white sample. The values of R, G, B of the black sample are subtracted from the values of R, G, B of the white sample and from the desired values of the measured sample. Then the values of R, G and B of the measured sample are divided by the corresponding values of the white calibration sample and multiplied by the maximum value of white. This means that for each image pixel, the calibrated value of R _{cal cal} values of R is calculated by the following formula:

R _cal = 255 · (RR _black ) / (R _bel -R _black ).

In this formula, R _black is the R pixel value of the black sample, and R _bel is the R pixel value of the white sample. Calibrated values for the values of B and G are calculated in a similar way. This correction takes into account the deviations of the photosensitivity of the pixels and the change in the light intensity as a function of optical geometry.

Optionally, the values of R, G and B can be adjusted with respect to time-dependent changes in light intensity. For this, for example, a parallel white strip can be applied to the sample. For calculation purposes, the sample and the white strip are mentally divided in the direction of the longitudinal axis of the sample into a large number of sections. For each section of the sample, the average values of R, G and B are determined, i.e. R _cf. , G _cf. and _Cf. Similarly, for each section of the strip, the average values of R, G and B are determined, i.e. R _bel-Wed , G _bel-Wed and B _bel-Wed Then, the corrected value of R is calculated, i.e. R _core for each plot of the sample according to the following formula:

R _cor = 255 · (R _cf / R _bel-cf ).

G _cor and B _cor are calculated in the same way.

The most common colorimetric data systems were installed by the Comission International de l'Eclairage (CIE), for example, CIELab (L *, a *, b *), CIEXYZ (X, Y, Z) and CIELuv (L * , u *, v *). These systems take into account the sensitivity of the human eye. The R, G, and B values measured by the camera on the CCD can be converted to the L *, a *, b * values of the CIELab system.

As a mathematical model, you can choose any model known to specialists in this field. Examples are given by HRKang, Color Technology for Electronic Imaging Devices, SPIE Optical Engineering Press, 1997, chap. 3 and 11, and US Pat. No. 5,850,472. A model may be linear or non-linear. As an example of a nonlinear model, a polynomial of the 2nd degree with 10 parameters can be given, or a polynomial of the 3rd degree with 20 parameters. A linear model is preferred. It is more preferable to use a linear model having 4 parameters.

As an example of a linear model with 4 parameters, consider the following model, where the measured color signals of the calibration colors, in this case, the data R, G and B, are converted to colorimetric data, in this case, the CIELab data:

L _i * = c ₀ + c ₁ R _i + c ₂ G _i + c ₃ B _i

a _i * = d ₀ + d ₁ R _i + d ₂ G _i + d ₃ B _i

b _i * = e ₀ + e ₁ R _i + e ₂ G _i + e ₃ B _i

where R _i , G _i , B _i are the measured signals, and L _i *, a _i * and b _i * are the colorimetric data of the calibration color i.

To calculate 12 model parameters c ₀ -c ₃ , d ₀ -d ₃ and е ₀ -e ₃ based on the measured RGB data and the known CIELab (colorimetric observer according to the CIE 1964 standard) calibration colors, linear regression is used. These model parameters are used to convert the measured RGB data of the selected color to CIELab data.

An example of a non-linear polynomial of the 3rd degree having 20 parameters is:

L _i * = c ₀₊ c ₁ R _{i +} c ₂ G _{i +} c ₃ B _{i +} c ₄ R _i ²⁺ c ₅ G _i ²⁺ c ₆ B _i ²⁺ c ₇ R _i G _{i +} c ₈ R _i B _{i +} c ₉ G _i B _{i +} c ₁₀ R _i ³ + c ₁₁ G _i ³ + c ₁₂ B _i ³ + c ₁₃ R _i ² G _i + c ₁₄ R _i ² B _i + c ₁₅ G _i ² R _i + c ₁₆ G _i ² B _i + c ₁₇ B _i ² R _i + c ₁₈ B _i ² G _i + c ₁₉ R _i G _i B _i

a _i * = d ₀₊ d ₁ R _{i +} d ₂ G _{i +} d ₃ B _{i +} d ₄ R _i ² + d ₅ G _i ² + d ₆ B _i ² + d ₇ R _i G _i + d ₈ R _i B _i + d ₉ G _i B _i + d ₁₀ R _i ³ + d ₁₁ G _i ³ + d ₁₂ B _i ³ + d ₁₃ R _i ² G _i + d ₁₄ R _i ² B _i + d ₁₅ G _i ² R _i + d ₁₆ G _i ² B _i + d ₁₇ B _i ² R _i + d ₁₈ B _i ² G _i + d ₁₉ R _i G _i B _i

b _i * = e ₀ + e ₁ R _{i +} e ₂ G _i + e ₃ B _i + e ₄ R _i ² + e ₅ G _i ² + e ₆ B _i ² + e ₇ R _i G _i + e ₈ R _i B _i + e ₉ G _i B _i + e ₁₀ R _i ³ + e ₁₁ G _i ³ + e ₁₂ B _i ³ + e ₁₃ R _i ² G _i + e ₁₄ R _i ² B _i + e ₁₅ G _i ² R _i + e ₁₆ G _i ² B _i + e ₁₇ B _i ² R _i + e ₁₈ B _i ² G _i + e ₁₉ R _i G _i B _i

To calculate 60 model parameters c ₀ -c ₁₉ , d ₀ -d ₁₉ and e ₀ -e ₁₉ , linear regression is used based on the measured RGB data and the known CIELab calibration color data. These model parameters are used to convert the measured RGB data of the selected color to CIELab data.

Despite the foregoing, when calculating model parameters, calibration colors close to the selected color can be assigned a larger weight. In the case of the above example of a linear model with 4 parameters, this means that during the linear regression, each calibration color is assigned a weight coefficient depending on the distance in the RGB color space between the desired calibration color and the selected color. In the linear regression procedure, the following sum of squares is minimized:

- расчетное значение для L*_i, a*_i или b*_i в зависимости от преобразования RGB в CIELab.where w _i is the weight coefficient, y _i is L * _i , a * _i or b * _i depending on the spectral measurements, and

- the calculated value for L * _i , a * _i or b * _i depending on the conversion of RGB to CIELab.

равно c₀+c₁R+c₂G+c₃B (см. выше), и w_i равен ((R_i-R)²+(G_i-G)²+(B_i-B)²)^-2, то эту сумму можно выразить следующим образом:If

is equal to c ₀ + c ₁ R + c ₂ G + c ₃ B (see above), and w _i is equal to ((R _i -R) ² + (G _i -G) ² + (B _i -B) ² ) ^-2 , then this amount can be expressed as follows:

Where

n is the number of calibration colors

R, G, B - measured signals of the selected color.

Alternatively, calibration colors close to the selected color can be used for interpolation.

If necessary, for signals measured for black, white and gray, it is possible to balance the gray according to the formula R = G = B = f (L *) or a comparable value for L * in another colorimetric system. This gray balancing is described by H.R. Kang, Color Technology for Electronic Imaging Devices, SPIE Optical Engineering Press, 1997, chap. 11.

Using an image processing program such as the Optimas ^® computer program commercially available from Media Cybernetics or the Image ProPlus ^® program available from the same company, individual particles can be identified by recognizing differences in illumination relative to the background of the particles. These particles can be, for example, metallic dyes or groups of metallic dyes. After identifying the particles, you can use the image processing program to determine the number of particles and image parameters, for example, particle size, particle shape, length of the shortest and longest axis, as well as the values of R, G and B particles. This data, optionally, can be averaged over a portion of the strip or, if necessary, over a portion of a larger size.

Data determined from images can be used, for example, to search for a coating formula that provides consistent surface coverage. For this, the measured data can be compared with the data from the database of color formulas.

To increase the coverage range of the image forming apparatus, the light source may optionally comprise a set of mirrors. It has been found that using mirrors in an appropriate configuration can increase the coverage range to about 90 degrees and even more.

Although the light source may be a constant light source, it is preferable to use a pulsed light source to minimize energy use. When using a constant light source, the appropriate exposure time of the camera should be set. Suitable light sources are, for example, tungsten halide lamps or xenon lamps.

According to a preferred embodiment of the invention, the light source comprises a diffuser housing from which light exits through the slit. The extended side of the slit is located essentially parallel to the surface of the sample, and the short side is almost perpendicular to the surface of the sample. In this configuration, a photosensitive device can be used to control the light output. According to a preferred embodiment of such a diffuser, the gap is bounded by a substantially horizontal wall on the inside of the diffuser. Thus, the light intensity on the sample surface is a function of the flop angle. At positions of a smaller angular distance from the diffuser, the light intensity is less than at large angles.

According to another preferred embodiment of the device according to the present invention, the spectral distribution of light changes in the image as a function of position on the sample, for example, through the use of different light sources or a set of filters, diffraction grating or prism. This allows you to increase the volume of independent measurement data and, thus, increase the accuracy of color reproduction. For this, it is possible, for example, to change the light from a light source or to change the spectral distribution of light immediately before it enters the image forming apparatus. Preferably, the spectral distribution of the illumination changes perpendicular to the change in optical geometry.

To exclude any influence of ambient light, the device according to the invention preferably comprises a casing.

The present invention provides a surface estimation method, according to which, as described above, in general, the detected light interaction is the reflection of light on a sample. However, if the sample is transparent or translucent, then the detected interaction of light may be a transmission of light. In this case, the sample is placed between the image forming apparatus and the light source.

You can use flat samples. However, if necessary, curved samples can also be used to study flop behavior.

The invention is further described and illustrated with reference to the drawings.

Figure 1 presents a diagram of the configuration of the registration according to the invention;

Figure 2 represents the registration in the configuration depicted in figure 1;

Figure 3 presents a diagram of an alternative configuration according to the invention;

Figure 4 presents a graph of the dependence of the flop angle on the position during registration in the configuration depicted in figure 3;

5 shows a third alternative configuration according to the invention;

6 shows a fourth alternative configuration according to the invention;

Figure 7 shows the change in the filtered wavelength within the sample, the image of which is formed by the device shown in Fig.6;

On Fig presents a sample with a parallel reference sample for measuring gloss.

Figure 1 shows the device 1 according to the present invention, containing a light source 2, a camera 3 on the CCD as a recording device having a viewing angle (ranging from the viewing direction 4 of the first external end closest to the light source to the viewing direction 5 of the second the outer end of the coated sample 6 is located under the camera 3. The light source 2 is a linear source parallel to the surface of the sample. The light source 2 is located outside the direct field of view of the camera 3 on the CCD. The line between the light source 2 and the point where observation board 4 intersects with sample 6, sets the first direction of illumination 7, which is reflected by sample 6 in the direction specified as the first direction of reflection 8. Similarly, the line between the light source 2 and the point where observation direction 5 intersects with sample 6 defines the second direction of illumination 9, which is reflected by sample 6 in the direction specified as the second direction of reflection 10. In the figure, the outer flop angle Θ ₁ is the angle between the first direction of observation 4 and the first direction of reflection 8, and the outer flop goal Θ ₂ is the angle between the second direction of observation 5 and the second direction of reflection 10. The range of angles between Θ ₁ and Θ ₂ can be up to about 90 degrees.

Figure 2 shows the image registered in the embodiment shown in figure 1. This is an image of a sample coated with metallic paint. This figure shows how the brightness changes with the flop angle. The image in figure 2 also shows the change in graininess along the sample, which is also perceived by the human eye.

Figure 3 shows an alternative configuration according to the invention, where the range of the flop angle is increased by using the mirror 11. The mirror 11 is positioned so that it reflects the part of the sample adjacent to the right to the outer end of the range of camera angles farthest from the source light 2. In the image observed by the camera 3, the coverage area closest to the light source 2 is replaced by the extension of the coverage area on the other side. From right to left, the registration shows a coverage area from Θ ₃ to Θ ₅ , and then a coverage area from Θ ₄ to Θ ₂ . Scope from Θ ₁ to Θ _{4 is} no longer visible on registration.

Figure 4 shows the dependence of the flop angle on the position in a configuration similar to the configuration shown in figure 3, with position 0 located directly below the camera. The part of the sample reflected from the mirror covers part from about 20 mm to 25 mm.

Optionally, the sample can be recorded in two (or even more) separate parallel strips, one with a coverage area expanded with a mirror, and the other without it. Thus, it is possible to register a complete enlarged coverage area ranging from Θ ₁ to Θ ₂ and a coverage area ranging from Θ ₅ to Θ ₃ . If Θ ₅ is Θ ₂ , then the closed range from от ₁ to Θ _{3 is} covered.

Figure 5 shows a configuration similar to the configuration shown in figure 1, in which the camera 3 is located directly above the sample 6, and the light source 12 contains a standard pulsed light source 13, available under the trademark Metz 45CT-1. The pulsed light source 13 comprises a transparent side 14. On the transparent side 14, the pulsed light source is connected to the flat upper side 15 of the diffuser 16 containing the semi-cylindrical part 17. This flat side 15 is open at the junction with the transparent side 14 of the pulsed light source 13. The inner side of the diffuser 16 is white coated. Where it does not coincide with the transparent side 14 of the pulsed light source 13, the flat side 15 is covered by a horizontal wall 18, the inner side of which is provided with a white coating. The edge between the outer end of the horizontal wall 18 and the semicylindrical part 17 is provided with a vertical slit 19 extending across the width of the diffuser 16.

When the pulsed light source 13 flashes, the light is scattered by reflective coatings on the inside of the diffuser 16. A portion of the light is reflected by the reflective layer of the horizontal wall 18 through the slit 19 to the portion of the sample 6 for which the image is to be formed. Figure 5 shows that in the external direction of the viewing range of the camera angle closest to the diffuser 16, the sample 6 is illuminated with a much smaller part of the horizontal reflecting wall 17 than on the external direction of the camera’s viewing, remote from the diffuser 16. Thus, the light density is function of the angular distance from the light source. The function may vary depending on the orientation of the slit with respect to the surface of the sample.

A fiberglass cable 20 connects the space surrounding the pulsed light source 13 with a photosensitive device 21, which controls the time interval during which the flash of the pulsed light source 13 occurs. A part of the light scattered in the diffuser 16 is transmitted through the fiberglass cable 20 to the photosensitive device 21. Photosensitive device 21 measures the amount of light transmitted through fiberglass cable 20. When a predetermined amount of light passes through fiberglass cable 20, photosensitivity The flash unit 21 blocks the pulsed light source 13. Thus, it is possible to guarantee that each flash gives exactly the same amount of light.

Fig. 6 shows another alternative embodiment of the device corresponding to the invention, comprising an image forming apparatus 3 and a light source 2. A sample 6 is located below the image forming apparatus 3. Between the light source 2 and the sample 6 there is a set of filters and a diffraction grating or prism 24 The spectral distribution of illumination changes in one image as a function of position on the sample. 7 shows the result of a preferred embodiment, where the spectral distribution of illumination changes perpendicular to the change in optical geometry.

On Fig shows that the gloss behavior of the sample can be characterized as a function of optical geometry. This particular example shows, in one image, the difference between the high gloss sample 25 and the low gloss sample 26.

Claims (9)

1. A method for recording goniochromatic characteristics of a paint film using a device containing an image forming apparatus for detecting light reflection of a paint film, a light source and a sample area for receiving a sample with a paint film to be examined, characterized in that the image forming device, the light source and the region of the sample is arranged in a triangular configuration for recording in one image at least one of the goniochromatic the characteristics of the paint film as a function of the continuous range of flop angles between the direction of reflection and the direction of observation. 2. The method according to claim 1, characterized in that the range of flop angles is more than 40 °, preferably more than 50 °. 3. The method according to claim 1 or 2, characterized in that the light intensity varies over the coverage area of the image forming apparatus. 4. The method according to claim 1, characterized in that the light source is a linear source. 5. The method according to claim 1, characterized in that the image forming apparatus is a CCD camera. 6. The method according to claim 1, characterized in that the device comprises a set of mirrors, expanding the coverage area of the image forming device. 7. The method according to claim 1, characterized in that the light source is a pulsed light source containing a light output control unit. 8. The method according to claim 1, characterized in that the device contains a set of filters and a diffraction grating or prism for changing the spectral distribution of lighting or the spectral distribution of light entering the recording device as a function of position on the sample. 9. The method according to claim 1, characterized in that the registered goniochromatic characteristic is the brilliance of a paint film.

Images