Basics #05 “Measuring color”
In general, there are probably few people who have ever practiced "measuring color." Even among art university professors who specialize in color, there are few who have a colorimeter in their laboratory, and mechanical color measurement seems to be the work of only a few people at specialized companies. Specialized equipment is a special precision instrument, and because there is little demand for it, it can be said to be expensive and difficult to obtain. On the other hand, products that can easily measure color using smartphone apps have emerged and are attracting attention.
So this time, I would like to think about "measuring color."
1. Purpose of color measurement
I think there are two main purposes for color measurement. Firstly, the purpose is to accurately understand the colors in order to produce the same color, and the second purpose is to understand the color distribution and the range of color usage in the market. I think there are two main reasons.
1) Mechanical color measurement
When manufacturing industrial products, it is necessary to produce exactly the same color. I believe that Japanese industrial products are extremely reliable in terms of color management. For information on the machines that measure color and their functions, the website of Konica Minolta (*or the company that manufactures color difference meters, etc.) is probably the easiest to understand. It is not my responsibility to explain it here, so please refer to it.
When measuring using a machine, large machines are more stable and accurate, but I think they have sufficient functionality even for simple color measurements. It is best to choose one based on the accuracy required for your work. The Japanese Industrial Standards indicate the range of acceptable color differences depending on the field. This is because the technology that allows for exactly the same color differs depending on the field.
- Specify the following areas of content:
nickname | Range of color difference ⊿E * | Degree of perceived color difference | Examples established in standards and regulations |
---|---|---|---|
Unevaluable area | 0~0.2 | Even specially calibrated color measurement instruments are within the margin of error and cannot be discerned by humans. | |
discrimination limit | 0.2~0.4 | The limit of what a trained person can reproducibly identify within the reproducibility of a well-calibrated colorimetric instrument. | JIS L 0804 Gray scale for discoloration JIS L 0805 Gray scale for contaminated colors standards etc. |
AAA class tolerance | 0.4~0.8 | In terms of reproducibility of visual judgment, there is a limit to setting strict standards for allowable color difference. | The strictest standards as agreed between the parties, etc. |
AA class tolerance | 0.8~1.6 | A level at which a slight color difference can be perceived when comparing adjacent colors. Allowable color difference range including errors between general colorimetric instruments | Defense Agency standards, Metropolitan Police Department standards, relatively strict internal standards and regulations for shipping inspections, etc. |
A class tolerance | 1.6~3.2 | The level of color difference is almost unnoticeable when comparing color distances. Levels that are generally considered to be the same color | JIS Z 8721 <Munsell color system color developer color sample>, etc., allowable color difference range by visual judgment between general standard color samples and sample colors, etc. |
B class tolerance | 3.2~6.5 | A range that can be treated as the same color at the impression level. In the paint and plastic industries, complaints may arise due to different colors. | Tolerable color difference commonly used in color management Allowable range of colors that are considered to be the same color between different color materials and material textures |
Class C tolerance | 6.5~13.0 | Color difference equivalent to 1 degree of JIS standard color chart, Munsell color chart, etc. | JIS D 0202 General rules for coating films for automobile parts JIS S 6005 Mechanical Pencil Pen JIS S 6006 Pencils, colored pencils, and the pens used for them JIS S 6007 blackboard JIS S 6016 stamp pad JIS S 6020 Vermilion JIS S 6028 watercolor paint standards etc. ![]() |
D class tolerance | 13.0~25.0 | The difference in color is such that it can be distinguished by subdivided systematic color names, and if it exceeds this level, it will give the impression of a different color name. | JIS E 3701 Level crossing facilities - Safety colors JIS Z 8102 Color names of object colors (especially surface colors) JIS Z 9101 Safety colors/safety signs JIS Z 9102 Piping system identification JIS Z 9103 Safety color general matters standards etc. *With the advancement of internationalization, the need for safe signs that everyone can understand at a glance has increased, and it is desired that they be standardized internationally. :2005 (Safety colors and safety signs - General rules for the design of safety signs for industrial environments and guidance), JIS Z9103 (Safety colors - General matters), JIS Z 9104 (Safety signs - General matters) Revised. ![]() |
Nippon Denshoku Industries Co., Ltd.
There are many companies that create tools called limit color charts that allow errors up to this range. This method is also approved by JIS. We create a range of colors in the form of color samples, such as a range of brightness that is acceptable or a difference in saturation that is allowed within this range.
Recently, it has become common to express the colors of landscapes in numerical values, and some people have thought that values other than those numerical values should not be accepted for things such as building materials. The tolerance level for the difference is set, so you should keep this in mind.
2) Visual measurement
There are many fields that use color measurement using the human eye.
This is an application of the JIS visual perception measurement method, but for color measurement, the color appearance of fabric products is measured in a box covered with a matte achromatic color with a lightness L* of approximately 45 to 55. There is a company called. Since the color of the product appears to change when the color of the light source changes, this is a way to check whether the color looks the same under the light source lamp environment used in your store. This is probably due to the idea that the presentation to the customer should be in the best possible condition.
Also, when measuring colors outdoors such as in buildings, color measurements are performed using color samples under natural daylight. At architectural conferences, there are research presentations on the variation in measurements made by the human eye, so it can be said that this requires some skill. Visual measurement of color can be said to be a technique whose accuracy increases with practice.
For reference, let's take a look at the "observer" mentioned in "JIS 8723 Surface Color Visibility Comparison Method."
Observers need the ability to discern subtle color differences. Therefore, it is desirable to test using various color vision test charts. When an observer uses eyeglasses for vision correction, the eyeglass lenses must have uniform spectral transmittance in the visible spectrum. To avoid the effects of eye strain, do not look at pastel and complementary colors immediately after dark colors. When comparing bright, vivid colors, if the comparison cannot be made quickly, the observer must focus on the achromatic color in the peripheral vision for a few seconds before making the next comparison. If the observer performs the task continuously, the visual perception performance deteriorates significantly, so the observer must frequently take a break of several minutes during which he or she does not perform the comparison task.
Observers are asked to check their color vision and are cautioned against decreased ability due to fatigue. People tend to think that looking at colors is fun, but in reality, it can be very tiring if you keep staring at them and paying attention to the differences.
If you read the above warning, visual color measurement may seem inaccurate due to difficult conditions, but if you compare the color sample with the actual object, you will be able to see the actual color better. Masu. It is safest to measure visual perception by comparing with a color sample if you can look directly at the object. When you want to get a rough idea, it's easier to understand if you compare colors that are close to gray by applying gray, if you want to see a whitish color, use white, and if you want to see a blackish color, apply black.
![Measuring the red color of the torii gate in the grounds of Shinjuku Tenso Shrine/Using color samples published by the Japan Paint Manufacturers Association](/ja/column/img/fig0501.jpg)
2. The mystery of numbers and how they feel
![](/ja/column/img/fig0502.png)
For example, with the yellow-red hue called YR, if you lower the saturation (make it more grayish) and increase the brightness (make it brighter), it will appear closer to pink. Even though the numerical value of the hue is the same, people perceive it differently. The way colors appear in nature and the rules of the color system do not exactly match.
In recent years, there has been an interesting movement in ``expressing things numerically.'' It seems that the ``mechanical colorimetric values'' are taking on a life of their own. Measuring with a machine allows you to finely expand the minute saturation. For example, using Munsell's values, it would be difficult to treat colors with a saturation of around 0.1 or 0.2 as chromatic colors from the perspective of range. Even though the saturation increases when the area is made larger, I think it is better to think of it as a situation where it looks like a grayish color.
If you look at the various things that come with expressing colors numerically, you may start to think that colors are a pain or that there is no need to measure colors. However, it is only by properly measuring colors that we can learn about the characteristics of ``perceived deviations'' and ``changes in the natural appearance of pigments and dyes.''
3. Utilization of color measurement data
By measuring colors and compiling the results into a hue, lightness, and saturation table, you can see which colors are concentrated, that is, which colors are used more frequently. These data are used in the color codes used in sales analysis in the distribution industry.
![Utilizing color measurement data](/ja/column/img/fig0503.jpg)
![Utilizing color measurement data](/ja/column/img/fig0504.jpg)
For example, the table and graph shown here measure and organize color trends proposed by multiple organizations for ladies in the fall/winter of 2016. This has the advantage of being easy to understand, as you can see a list of what hues and tones (color tones such as pale, dark, flashy, plain, etc.) are increasing.
When you measure colors, you can feel that various situations are buried in the data, such as understanding the current state of how colors are used, whether there are any deficiencies, and predictions of color expectations. If you compare colors over time and continue to accumulate data, you will find that some colors have been loved for many years.
4. Expansion of “measuring color”
I think the convenience of representing colors numerically will continue to be used more and more as a common language for people with various visual characteristics. As I mentioned at the beginning, the ability to easily measure color on smartphones and other devices is progressing, so I think it will become more commonplace. I hope that people will think about how to use it effectively and make full use of it while taking into account the fact that it can be seen differently and the range of permissibility depending on the field.
August 31, 2016
Text by Japan Color Design Institute