JPH0643920B2 - Method and apparatus for determining color tone of natural stone material - Google Patents

Description

TECHNICAL FIELD The present invention relates to a method and an apparatus for discriminating color tone, color unevenness (spots), etc. of a material from image data obtained from the surface of the natural stone material.

[Prior Art] Natural stone materials such as granite and marble are used as building finishing materials for inner and outer walls of buildings and the like. For example, 1 meter square granite plates are attached to the wall one by one and finished.

The surface of the natural stone material has a different color tone depending on the place of origin and the part of the rock, and also has different spots, scratches, stripes, and contamination with impurities. Therefore, if the cut-out plates are not uniform, color unevenness will occur on the finished wall surface, and it will not be possible to obtain a naturally uniform finished surface.

Conventionally, in order to eliminate variations in the color tone of natural stone materials such as granite and marble, especially architectural finishing materials, a visual inspection was conducted to remove each non-standard board from the standard sample one by one in comparison with the standard sample. The law was adopted. In addition, when a more uniform finished surface is required, all the finishing materials are actually laid in a large stone yard, and an inspector performs a visual inspection from the top of the yagura to exclude non-standard products and The items were rearranged to increase the homogeneity.

[Problems to be Solved by the Invention] However, since the conventional visual inspection relies on human vision, there are individual differences naturally, and when the natural stone material is actually attached to a wall, the owner, There were many discrepancies between designers and builders regarding the homogeneity evaluation of finished surfaces.

It is possible to judge the color tone of an object by using a color-difference meter that quantifies and quantifies the indices of optical brightness, hue, and saturation.In this color-difference meter, the surface of the subject is digitized. However, it is not possible to inspect the surface of the object having a large area as a whole and to determine the color tone.

The present invention is based on the above background,
The purpose is to obtain an index that accurately represents the color tone of the natural stone from the image data obtained from the surface of the natural stone material and quantify it. It is intended to provide a method and an apparatus for objectively discriminating spots and the like rather than subjectively.

[Means for Solving the Problems] The above-described problems can be solved by the color tone determination method and apparatus thereof according to the present invention.

That is, the color tone determination method of the present invention is characterized by including the following steps.

(B) A process of taking an optical image of the surface of a natural stone material with a color imaging device and converting each pixel of the image into an electrical signal corresponding to the densities of red, green, and blue. The process of calculating the red density value R, the green density value G and the blue density value B from the signal, and the brightness I of the pixel defined as the average value of these values. (C) The red density value R of each pixel , The green density value G, the blue density value B, and the lightness I data, the process of classifying the mineral components of each pixel based on the mineralogy setting value. (D) All pixels of the entire image or partial image taken The average value AI of the brightness and the area ratio of each mineral component, which is the ratio of the number of pixels classified into each mineral component to the total number of pixels, are calculated, and preferably a white mineral component, Red-based mineral composition to area ratio of plagioclase-based mineral composition There is a process of calculating a reddish correction value which is a ratio of the area ratio of mineral components mainly composed of potassium feldspar to obtain an index used for the color tone determination of the natural stone material. In a preferred embodiment of the present invention, the mineral component of each pixel is The classification can be determined based on the following criteria. That is, the threshold value for the lightness I preset from the standard stone is
It is defined that IM $, IP $, and IK $ are in ascending order, and the threshold values for the red concentration value R preset from the standard stone are RK $ and RF $ in ascending order.

This criterion will be specifically described with reference to the accompanying drawings showing R, G, B, and I spaces.

FIG. 2 shows a color cube with Cartesian coordinates R, G and B and a diagonal axis I.

Pixels belonging to colored minerals satisfy the following conditions.

In the space of I ≦ IM $ (R, G, B), it is shown by the subfield of FIG.

Pixels belonging to plagioclase meet the following conditions:

(R <RF $ AND IP $ ≦ I <IK $) OR
In the space of IK $ ≦ I (R, G, B), it is shown by the subfield of FIG.

Pixels belonging to potassium feldspar satisfy the following conditions.

(RK $ ≦ R <RF $ AND IM $ <I <IP $)
In the space of OR (RF $ ≦ R AND IM $ <I <IK $) (R, G, B), it is shown by the subfield of FIG.

Pixels belonging to gray quartz satisfy the following conditions.

In the space of R <RK $ AND IM $ ≦ I <IP $ (R, G, B), it is shown by the subfield of FIG.

In a preferred embodiment of the present invention, not only the brightness AI,
Color unevenness can be further reduced by comparing the reddish correction values and determining the color tone. In this case, colored minerals,
The area ratio of each mineral component, which is the ratio of the number of pixels classified as plagioclase, potassium feldspar, and gray quartz to the total number of pixels, is defined as MMD, PLD, KFD, and GQD, and the reddish correction value BI Is defined as follows.

BI = KFD / PLD In a preferred aspect of the present invention, the average brightness AI, the area ratio of each mineral component, and the reddish correction value can be calculated for a partial image without processing the entire captured image. .

A natural stone material color tone determination apparatus according to the present invention is characterized by including the following means.

An optical image of the surface of a natural stone material was taken with a color imaging device
From the electric signals corresponding to the red, green and blue densities of the respective pixels of the image obtained by the conversion, the red density value R, the green density value G and the blue density value B, and these values of Calculation means for calculating the lightness I of a pixel defined as an average value From the data of the red density value R, the green density value G, the blue density value B, and the brightness I of each pixel, the mineral component of each pixel is subjected to mineralogy. Means for classifying based on the statically set value Calculate the average value AI of the brightness of all pixels in the photographed whole image or partial image and the area ratio of each mineral component which is the ratio of the pixels classified into each mineral component And preferably,
Further, a redness correction value BI, which is the ratio of the area ratio of the mineral component mainly composed of potassium feldspar, which is a red mineral component, to the area ratio of the mineral component mainly composed of plagioclase, which is a white mineral component, is calculated. Then, the index calculation means for obtaining the index used for the color tone determination of the natural stone material In a preferred embodiment of the device according to the present invention, the quality of the natural stone material is determined by comparing the obtained index with a preset limit range. Judgment means, output means for outputting a judgment result, and an optical image of the surface of the natural stone material is photographed and converted into an electric signal corresponding to the red, green and blue densities of the respective divided pixels. And a color imaging device.

A block diagram of an example of a measuring apparatus according to the present invention is shown in FIG.

In this example, a small-sized color television camera (ITV) as an imaging device, an image processor that functions as a calculation unit, a classification unit, and an index calculation unit, a host computer that performs various calculation commands, controls, and determinations, a printer, and It is composed of an output device such as a CRT and a color television monitor.

[Operation] The operation of the color tone determination method for a natural stone material and the apparatus thereof according to the present invention configured as described above will be described below by using the method as a representative.

In the color tone determination method of the present invention, first, an optical image of the surface of a natural stone material is photographed by a color imaging device and converted into an electrical signal corresponding to the image data. In this image pickup device, an image is displayed, for example, 512 × 512 pixels (pixels).
And each pixel is converted into an electric signal corresponding to the densities of red, green and blue. Here, red, green and blue are the three primary colors of light.

Next, from the obtained electric signal, the red density value R, the green density value G and the blue density value B, and the brightness I of the pixel defined below are calculated.

In the process of I = (R + G + B) / 3 (C), the mineral component of each pixel is mineralogically determined from the data of the red density value R, the green density value G, the blue density value B and the lightness I of each pixel. Classify based on the set value. this is,
Natural stones are generally not uniform in composition and are aggregates of several or dozens of mineral grains.For example, granite is mainly
It is composed of colored minerals (black), plagioclase (white), potassium feldspar (pink or red) and gray quartz (gray) minerals, each of which has a relatively large size of 1 mm or more and has its own unique color. There is. Therefore, by comparing the measured color information of each pixel with the color information of a typical standard mineral, it is possible to determine which mineral each pixel belongs to.

In the next step (d), the average value AI of the brightness of all pixels and the area ratio of each mineral component, which is the ratio of the pixels classified into each mineral component, are calculated for the photographed whole image or partial image. In a preferred embodiment, the red is the ratio of the area ratio of the mineral component mainly composed of potassium feldspar which is the red mineral component to the area ratio of the mineral component mainly composed of the plagioclase which is the white mineral component. The taste correction value BI is calculated to calculate the index used for determining the color tone of the natural stone material. By using the reddish correction value BI in addition to the average value AI of lightness as an index used for color tone determination, the accuracy of color tone determination can be increased.

Here, the average brightness AI of the captured image is an index indicating optical brightness, and the color tone of the natural stone material that is the subject is roughly determined using the index. Furthermore, the calculated reddish correction value is introduced because it is optically bright when there is a lot of white, but it tends to appear dark to the human eye when there is a lot of white (red) even if there is a lot of white. This is a correction term, and by introducing this correction term, the error caused by quantifying and quantifying the color tone by eliminating human subjectivity can be suppressed to a low level.

The limit range of the index is set in advance for the material having the desired quality, and this is compared with the index of the subject,
Determines the color tone and color unevenness of natural stone materials.

[Effects of the Invention] The following effects can be obtained by the measuring method and apparatus of the present invention.

(A) By the method for determining the color tone of a natural stone material according to claim 1, an index that accurately represents the color tone of the natural stone is obtained and quantified and quantified. ), Etc., can be objectively determined without individual differences.

(B) In the method for determining a color tone of a natural stone material according to claim 2, a redness correction value considering human visual characteristics is introduced without using only optical brightness as an index of color tone, and a correspondence with visual observation is provided. I can get it well.

(C) In the color tone determination device according to the eighth aspect, the calculation and control can be performed by a computer, and the color tone and the like can be determined in a short time.

(D) In the color tone determination device according to the tenth and eleventh aspects, by integrating a color television camera and a means for performing various functions, a portable color tone determination device can be obtained, which is a simple measuring instrument.

(E) In the color tone determination device according to any one of claims 8 to 11, the obtained image data is stored in a memory, the surface of the natural stone material is reproduced from this data by using computer graphics, and further, a construction finishing material is arranged. You can perform different simulations.

(F) In the method according to claim 5, the entire surface is divided, the index is calculated for each partial surface, and the color tone is determined for each partial surface. Therefore, unevenness (variation) on the entire surface is detected. Further, by increasing the accuracy, it is possible to determine the direction and width of scratches and cracks in the natural stone material.

(G) The color tone determination method of the natural stone material according to claim 1 can be applied to a painted surface similar to the natural stone material.

(H) Unlike the conventional method, there is an advantage that it is not necessary to manually move a heavy natural stone material only for inspection.

[Examples] The present invention will be specifically described with reference to Examples.

Using the measuring apparatus shown in FIG. 1, the color tone of the building exterior wall finishing material of Canadian granite Caledonia was determined according to the present invention.

First, the surface of a building exterior wall finishing material in Canadian granite Caledonia is photographed with a color image pickup device (CCD color television camera), the image is divided into 512 × 512 pixels, and the red, green, and blue densities of each pixel are divided. Is converted into an electric signal corresponding to, and the obtained image data is loaded into the image memory.

Then, in the image processor, from the obtained data,
The red density value R, the green density value G and the blue density value B, and the lightness I of the pixel defined as the average value of these values were calculated. The threshold value for the lightness I and the threshold value for the red density value R were preset from the standard stone as follows.

IM $ = 36 IP $ = 77 IK $ = 96 RK $ = 79 RF $ = 99 The mineral components of the portion corresponding to each pixel were classified using the following criteria.

Pixels belonging to colored minerals satisfy the following conditions.

I ≦ IM $ Pixels belonging to plagioclase meet the following conditions.

(R <RF $ AND IP $ ≦ I <IK $) OR
IK $ ≦ I Pixels belonging to potassium feldspar satisfy the following conditions.

(RK $ ≦ R <RF $ AND IM $ <I <IP $)
OR (RF $ ≦ R AND IM $ <I <IK $) A pixel belonging to gray quartz satisfies the following condition.

R <RK $ AND IM $ ≦ I <IP $ Average value AI of brightness of all pixels and area ratio of each mineral component which is a ratio of pixels classified into each mineral component MMD, PL
D, KFD, and CQD, and a reddish correction value BI (= KFD / PLD), which is a ratio of the area ratio of the red mineral component to the area ratio of the white mineral component, were calculated.

The results are shown in FIG.

In this figure, the vertical axis shows the redness correction value BI, and the horizontal axis shows the average brightness A.
I is shown, and the measurement result of each Caledonia plate is shown by a black dot. The range preset for the reference Caledonia plate as allowed by visual inspection is the rectangular inner region of the figure.

The test Caledonia plates contained in the rectangular inner region showed good results by visual inspection, and the other test plates did not show good results by visual inspection. From this, it was found to have a good correspondence with the visual inspection.

[Brief description of drawings]

FIG. 1 is a block diagram of an example of a measuring device according to the present invention, FIG. 2 is an explanatory diagram for explaining a color cube of rectangular coordinates R, G and B, and a diagonal axis I, and FIG. 4 is an explanatory view showing a partial body belonging to plagioclase, FIG. 5 is an explanatory view showing a partial body belonging to potassium feldspar, and FIG. 6 is a partial body belonging to gray quartz. Illustration,
FIG. 7 shows the measurement results of Canadian granite Caledonia.

Claims (11)

[Claims] 1. An optical image of the surface of a natural stone material is photographed by a color image pickup device and converted into an electric signal corresponding to the densities of red, green and blue for each pixel of the image, and the obtained electric signal is obtained. Then, the red density value R, the green density value G and the blue density value B, and the lightness I of the pixel defined as an average value of these values are calculated, and the red density value R and the green density of each pixel are calculated. From the data of the value G, the density value B of blue, and the lightness I, the mineral components of each pixel are classified based on the mineralogy setting value, and the average value AI of the lightness of all the pixels of the whole image or partial image taken is calculated. A method for determining a color tone of a natural stone material, comprising calculating an area ratio of each mineral component, which is a ratio of pixels classified into each mineral component, and calculating an index used for determining the color tone of the natural stone material. 2. A reddishness correction value which is a ratio of an area ratio of a mineral component mainly composed of potassium feldspar which is a red mineral component to an area ratio of a mineral component mainly composed of plagioclase which is a white mineral component. The method according to claim 1, wherein is further used to determine the color tone of the natural stone material. 3. The method according to claim 1 or 2, wherein each pixel is classified into mineral components based on the following criteria (provided that the threshold value for lightness I preset from standard stone is in ascending order: IM $, IP $, and IK $, and the threshold values for the red concentration value R preset from the standard stone are RK $ and RF $ in ascending order). (A) Pixels belonging to colored minerals satisfy the following conditions. I ≦ IM $ (b) Pixels belonging to plagioclase must meet the following conditions. (R <RF $ AND IP $ ≦ I <IK $) OR
IK $ ≦ I (C) Pixel belonging to potassium feldspar must meet the following conditions. (RK $ ≦ R <RF $ AND IM $ <I <IP $)
OR (RF $ ≦ R AND IM $ <I <IK $) (d) Pixels belonging to gray quartz satisfy the following conditions. R <RK $ AND IM $ <I <IP $ 4. The area ratios of mineral components of colored minerals, plagioclase, potassium feldspar, and gray quartz are MMD, PLD, and K, respectively.
The method according to claim 3, wherein the reddishness correction value BI is defined as follows, where FD and CQD are used. BI = KFD / PLD 5. The method according to claim 2, wherein the average brightness AI, the area ratio of each mineral component, and the reddishness correction value are calculated for a partial image obtained by dividing the photographed entire surface image into a plurality of parts. 6. The natural stones are granites, 1, 2, 3,
The method according to any one of 4 and 5. 7. The natural stone is marble, 1, 2, 3,
The method according to any one of 4 and 5. 8. A red density value R from an electric signal corresponding to the red, green and blue densities of each pixel of the image obtained by photographing an optical image of the surface of a natural stone material with a color image pickup device and converting the image. , A density value G of green, a density value B of blue, and a lightness I of a pixel defined as an average value of these values, and a red density value R, a green density value G, and a blue of each pixel. From the density value B and the lightness I data of, the classification means for classifying the mineral component of each pixel based on the mineralogy setting value, and the average value AI of the lightness of all the pixels of the photographed whole image or partial image, An apparatus for calculating a color tone of a natural stone material, comprising: an index calculation unit for calculating an area ratio of each mineral component, which is a ratio of pixels classified into each mineral component, to obtain an index used for determining a color tone of the natural stone material. 9. The index calculating means further comprises an area ratio of a mineral component mainly composed of potassium feldspar which is a red mineral component to an area ratio of a mineral component mainly composed of plagioclase which is a white mineral component. The color tone determination device according to claim 8, wherein a reddishness correction value BI, which is a ratio, is calculated and used for color tone determination of the natural stone material. 10. Judging means for judging the quality of the natural stone material by comparing the obtained index with a preset limit range,
The color tone determination device according to claim 8, further comprising an output unit that outputs a determination result. 11. A color image pickup device comprising an optical image of the surface of a natural stone material and converting the image into electric signals corresponding to the densities of red, green and blue for each of the divided pixels. , 9 or 10 any one of the color tone determination devices.