How did the problem occur?

The computer does not understand colors. Basically, they are just an adder that processes 0 and 1 according to our instructions. When we use 0 and 1 to reproduce colors on the computer, we actually created an RGB digital control system or CMYK analog control system to control various computer peripherals with color reproduction capabilities, such as scanners, monitors, printers, imagesetters, plate-making machine, etc.

  The RGB and CMYK control systems essentially use a mixture of three or four primary colors to produce the desired color. The strength of the signal of each component determines how many corresponding primary colors are used. When we use numbers to express RGB and CMYK colors, we just use numbers to reproduce the intensity of each component.

  When we reproduce color on a particular device, this system works very well. Unfortunately, when we send these same RGB and CMYK values ​​to different devices, they usually produce different colors. This is because RGB and CMYK generate electrical signals on computers instead of specific colors, and each device responds differently to these electrical signals.

  If you have ever watched TVs of different brands in an electronic store, you will see this actual phenomenon: although many display screens receive the same signal, they produce different colors. why? Because the RGB value is used to control the intensity of the electron beam to make the phosphor of the display generate heat, thereby emitting light, and each display responds differently to the signal. One of the reasons is that different manufacturers use different phosphors, and their responses to electron beams are different. But this is only part of the problem. Also, the degree of attenuation and aging of phosphors are also inconsistent. This attenuation will affect the ability to generate light. Another reason is that the user's own settings of the brightness and contrast of the display will also affect the color display. We can even say that each display's ability to reproduce color is unique.

  For some slightly different reasons, the same will happen to other RGB and CMYK devices we use. Different brands of scanners and digital cameras use different color filters. The color filtering capabilities of these filters will change with the use of time, and each product will use different light sources, different scanner light sources The spectral curve is different. And the light source of the environment of the digital camera during shooting is also ever-changing. CMYK has more variable factors than RGB. There are many formulas of inks, glazing waxes, and dyes, all of which will cause different colors. If you also take into account the paper factor, you will introduce another big variable factor, because different papers The way the ink is affected is very different.

  You can think of RGB and CMYK as different formulas that produce colors, and different RGB and CMYK values ​​are the ingredients. It's like cooking, the same ingredients with different chefs will make different flavors. The color is the same: yes, you can say with certainty that R255, G255, B50 will produce a yellow, but this yellow produces different colors on different devices. 

  RGB and CMYK are often referred to as device-specific or device-dependent color models, precisely because only specific devices can be given to predict the effects of colors. There are two meanings here: First, the same value will produce different colors on different devices, second, to get the same color on different devices, you must change the value. This is the most basic problem that color management solves. There are also many obvious problems that will arise from the first problem: when we send files from one device to another, the color has changed, so the color seen by the scanner is the same as what we see on the display The colors are inconsistent, and the colors of the printout also do not match the colors on the monitor.

  The color of the sample seen by the scanner is R247G160B91. But when we send the same value to the display, it becomes a little darker and the saturation increases. When we send the same data to the printer, it becomes darker and more saturated. 

Reference color space

  The color management system we use today actually uses two reference color spaces, called CIE XYZ (1931) and CIE Lab (1976). We have already learned about these two color spaces in more detail in the previous lecture. To understand how the color management system works, you only need to understand two things about these two color spaces:

  Lab is the mathematical equivalent of XYZ

  Whether the colors described by XYZ or Lab are defined according to our human feelings, instead of using controlled electrical signals to produce colors like those that produce colors. In other words, the colors defined by these two color spaces can be seen by a person with normal vision. This person we already knew in the previous lesson is called "Standard Observer 1931". As a result, the colors defined by the values ​​of XYZ and Lab are clear, rather than the color space models related to devices such as RGB and CMYK. Although the values ​​are known, the exact color cannot be predicted if the specific device is not specified.

Device characteristics file

  The device characteristic file provides us with the color device used to describe the behavior of colors. If this is an RGB device, its device characteristic file explains what color each RGB combination of this device reproduces. In order to facilitate understanding, we can simplify it, (but in fact Profile is very complicated, we will talk about this in later courses). We can think of the device's characteristic file as a bilingual dictionary of colors, one language is the actual color sensed in XYZ or Lab, and the other language is the RGB or CMYK values ​​associated with the device. The characteristic file of the device connects the control signal (RGB or CMYK value) of this device with the actual perceived color produced on it, that is, the clear Lab or XYZ value. 

to sum up

  The color management system is actually very easy to understand. If you remember a simple principle in your heart: an independent feature file makes the RGB or CMYK data have a clear color; if you want to keep the color consistent, you need to change the data in the file , And this requires two feature files. So it is a good habit to embed a feature file in your file, especially when you need to send this file to other people or want to keep it for a long time (remember that as your monitor uses time Increasing its color will also change). When you embed a characteristic file into an image file, you paste a description to the file, explaining what the color data contained in it contains in reality and does not change the data itself, so this It also dispels the concerns of many people, you can safely embed the correct feature file into the file, because this will not destroy the RGB or CMYK data in the file. When you ask the CMS to match colors on another device, you need to specify two characteristic files, one indicating where these values ​​come from and the other indicating where they are going.

  If there is no mosaic feature file, the colors in the file are just a bunch of data, and different devices can interpret them in different colors, as we saw in Figure 1. When we embed the characteristic file, CMS will say RGB247, 260, 91 of the scanner, RGB250, 175, 100 of the monitor, and RGB244, 192, 148 of the printer will produce the same LAB value of 79,19, 46 light orange.

  Complete color management applications such as Photoshop and Illustrator add another key point. They use a theoretical RGB color space that does not depend on any device, such as Adobe1998, Apple RGB, sRGB1966, etc., and the display is independent of the color conversion of the source device and the target device, and the data is converted by doing a quick internal conversion. Sent to the display card, so the color display on each independent display is also correct. In fact, the internal processing is still the same: the application first observes the source characteristic file (that is, the current working color space of the application), judges its actual value in the theoretical RGB, and then observes the target characteristic file (monitor) and judges that it should be What kind of RGB values ​​are used to reproduce this color, and then converted to the display through the graphics card.

  There are many branching topics about color management, such as the choice of Rendering Intents, but as long as you understand this simple rule in your mind-you need a Profile to describe color, you need two Profiles on two devices Match colors between times-you will find that color management increases your understanding of colors, saves you time, and reduces waste.