Fundamentals of Post-photography (I): Color and Color Mode

This paper is a series of digital photography post-processing courses, focusing on PS technology. ?

This chapter mainly introduces how to quantify the colors we usually see and what methods are available. These quantization methods have direct corresponding interfaces and modes in Photoshop. For example, Lab mode, CYMK mode, various RGB modes, color selector and HSB color panel in hue adjustment tool. Only by understanding these patterns can we know what the specific operation will be like.

Have many people encountered various basic problems in the process of learning PS, such as why people's skin color can be adjusted through Lab mode? What is a laboratory? Is it a laboratory model? Why is 16 bit image easier to use? What exactly is a mask? What is this paragraph? What is the relationship between channel and mask? Under what circumstances can it be converted? What does the mask have to do with the constituency? How do black masks and white masks work on layers? And what is the principle of layer blending mode, and the effect can only be tested?

In this case, the list of questions can be very long. If you simply recite the explanations and answers to these questions, you will find yourself in a painful cycle: recite, use, then forget, and then new questions come again. In fact, as long as you have a good understanding of some basic concepts of digital image processing (note that I am talking about concepts, not algorithms, so don't be intimidated, and algorithms have PS for you), you can understand the basis of the whole PS work: What is it dealing with? What is its main treatment? You will naturally understand a bunch of questions I have listed above, and there are still some questions you can't ask.

The concrete structure of this paper is as follows. We generally advise readers to read in order, but for basic or impatient readers, you can skip the previous content directly. If they find something difficult to understand, they can go back to the previous chapter to find the answer.

First of all, we will explain the concepts of color pattern and color space, so that readers can understand how a color in the real world is represented by the concept of quantity. And why to do color management, why to buy professional display equipment and correct colors.

We will go deep into the soul to answer a fundamental question for readers: what are the photos taken by our cameras and how do our photos represent this colorful world? Here, I will tell you why taking photos in RAW format will have more post-production space, and what HDR is.

Next, this paper will tell you the essence of PS image processing, various adjustments (contrast, brightness, color scale, color scale adjustment), and what the filter has done to your photos.

Through the above knowledge, readers will understand how the powerful selection function in Photoshop is realized, and the mathematical principle of layer mask and layer blending. Don't worry about your math knowledge. In fact, layer mask and mixed functions only use the simplest operations of addition, subtraction, multiplication and division, which are very simple. Starting from this part, I will cooperate with some practical cases of retouching, and impatient readers can also start with cases first. However, we strongly recommend that readers learn the principle first and then look at the actual cases, and verify the knowledge from the cases, so as to master it more effectively.

In our daily photography, we often hear a sentence: the color of this film is great! I wonder if readers have ever thought about how the colors we see are quantified? To understand this process, we have to mention two concepts: color mode and color space.

Note: The correct translation of color mode should be color mode, but I think it should be called color coding mode more accurately.

Before talking about color, we must first understand what light is. Light can be defined as electromagnetic waves in physics, and the light that people can see is called visible light, which is only a very narrow part of the frequency range of electromagnetic waves. The wavelength range of visible light is about 760nm to 380nm, which varies according to red, orange, yellow, green, blue, indigo and violet light, and the light colors in different wavelength ranges are also different. The white light we see is composed of many colors, and the reason why we can distinguish colors depends largely on the color components contained in the light. If the frequency of a beam of light is single, we say that the purity of color is higher. A color with a single frequency is called spectral color.

It is precisely because the purity of the light we usually shoot cannot be very pure spectral color, so the light entering the lens is usually composed of light with different wavelengths. For light with different wavelengths, the refractive index of the lens glass is different, which is an important reason for lens dispersion.

Color mode is a way to express color in the digital world. In the digital world, in order to express various colors, people usually divide colors into several components. Due to the different principles of color formation, color devices such as displays, projectors, scanners, etc., which synthesize colors directly through color light, are different from printers, printers, etc., which use pigments to generate colors. (The above is from Baidu Encyclopedia).

Common color modes are divided into two categories: additive color mode and subtractive color mode.

The simple metaphor of addition mode is projector mode, such as RGB mode, which is the most common addition mode. In this mode, white light is a mixture of red, green and blue light. In addition, YUV mode is also a common color mode, and YUV mode usually appears in video format.

The subtractive color mode is a simple printing machine mode. CYMK mode is common and seldom used in our image post-processing. However, if you need to send and print after saving, you need to save it as a picture in CYMK mode. JPEG format is a mode that supports CYMK.

Here we focus on the common color patterns:

1.RGB mode.

RGB mode, also known as tricolor mode, is an additive mode, which adds three colors of red (R), green (G) and blue (B) in different proportions to synthesize various colors of light. These colors define a color space. We define the red quantity as the X-axis, the green quantity as the Y-axis and the blue quantity as the Z-axis, thus obtaining a three-dimensional space model. This pattern can be visually represented by a color cube.

Then there is a question, why use RGB primary colors instead of other primary colors? Do these three frequencies of light have special physical significance? Actually, it's not. The answer is in our eyes. Therefore, the theory of red, blue and green tricolor light is not because these three wavelengths of light are special, but because our human eyes can only see these three wavelengths and give them colors.

2.HSL/HSB/HSV model

HSV/HSB uses hue (hue), saturation and lightness to represent color, and hue, saturation and value (lightness) to represent demonstration. A table can usually be represented by an inverted cone, as shown in the following figure:

Because it is an inverted cone, the lower the lightness, the smaller the color gamut. Visually speaking, decreasing the brightness of the picture will increase the saturation. This is also the reason why the saturation should be reduced appropriately when the brightness is reduced in post-processing.

HSL is similar to HSV, except that L stands for lightness. In HSL, when l is 100, it means white. But in HSV, when V is 100, it indicates the spectral color. If HSV is used to represent white, not only V should be equal to 100, but also color purity S=0. HSV mode is used in PS, so we won't discuss HSL in depth here.

Whether HSV/HSB or HSL mode, the color cone is a circle from top to bottom. One of the circles is the spectral color. This circle is our common color wheel. It is a very important knowledge to make an arbitrary straight line through the center of the circle, and the colors at both ends of the straight line complement each other.

Let's use an example to open the color picker of PS, as shown below:

Similarly, as a color adjustment tool in PS, it also adjusts the color based on HSB. (Note: Although the PS interface shows that the saturation is A and the lightness is I, it is exactly the same as the color purity and lightness in HSB), as shown in the following figure:

Note that in the above two tools, the value of hue (H) is-180 to+180. According to the concept of color wheel mentioned above, to obtain the complementary color of any hue value, just take a value with a difference of 180. That is, the hue value of+180 less than 0 is complementary color, and the hue value of 180 less than 0 is complementary color, which is very useful in our accurate color matching.

3.YUV mode

YUV is not commonly used in PS, but it is often used in video coding. Here is a brief introduction. Where Y stands for lightness, and UV stands for blue channel and red channel in chromaticity, as shown in the following figure:

The value of y is 0 to 1, and the larger the value, the brighter the color. The value of UV is -0.5 to 0.5, and the larger U, the closer to blue. The bigger the v, the closer it is to red.

Note: SIGMA Photo Pro6, the RAW processing software of Sigma, uses UV color plane to select colors.

This coding method is often adopted by TV and video. Human eyes are more sensitive to brightness perception than color perception. In video compression, the resolution of UV is often reduced to reduce the amount of data. The common YUV 4:4:4/YUV 4:2:2 is named according to the different resolution of UV.

The specific data compression methods are as follows:

YUV？ 4:4:4 sampling, each Y corresponds to a set of UV components.

YUV？ 4:2:2 sampling, every two Y shares a set of UV components.

YUV？ 4:2:0 sampling, every four Y shares a set of UV components.

From left to right, the amount of data decreases gradually, and the corresponding image quality also decreases gradually.

4. Laboratory mode

Lab mode is a color opposition space, L stands for brightness, with the range of 0~ 100, and A and B stand for color opposition dimensions. A* represents the component from green to red, and the value is-128~ 127-b* represents the component from blue to yellow, and the value is-128~ 127.

Lab color model is a color model based on physiological characteristics. But according to the principle of human vision, primates have two visual channels: red-green channel and blue-yellow channel. Most animals have only one passage at most. If someone lacks one of them, it is what we call color blindness. You can regard each color as a point in three-dimensional space (with three components: L, A and B) and calculate the Euclidean distance between them.

This value is called color vision distance. When the color vision distance is less than 30, the two colors are too close to distinguish, and when the color difference is greater than 120, the two colors are too different to form a strong contrast. The value of Δ e is also an important indicator of the display, which is used to measure the color accuracy of the display.

5.？ CYMK mode

CYMK is the subtractive color mode, CMYK represents the four colors used in printing, C represents cyan, M represents magenta, Y represents yellow, and K represents black. Because in practical application, it is difficult for cyan, magenta and yellow to form real black, and at most it is brown. So k black was introduced. The role of black is to strengthen the dark tone and deepen the dark color.

CYMK is the best printing mode. When PS is processed, RGB and HSV modes will eventually be used for calculation. Therefore, in the post-processing process, the more suitable working mode is to edit in RGB mode first, then print in CMYK mode, and then convert before printing, and add necessary color correction and cutting when converting.

Add some knowledge of photography other than visible light. Above 760nm, we call it infrared light, and below 380nm, it is ultraviolet light. Although we can't see the light with the naked eye, our camera can actually "feel" the light.

Let's talk about infrared ray first, because our camera can receive infrared ray, so we can take infrared ray photography. Just add an infrared filter in front of the lens to filter out visible light with shorter wavelength than infrared light. Because of this, most infrared filters look black. A very important parameter for purchasing infrared filters is wavelength, and the model is usually marked with IRxxx. The number after IR here is to let infrared light near this wavelength pass through the filter. You can choose according to your own needs when buying.

Wavelengths below 380nm become ultraviolet light, which is invisible to our naked eyes. The English word for ultraviolet ray is ultraviolet ray. Do you feel familiar when you see the word ultraviolet? That's right. Usually, the UV of the UV mirror we use refers to the UV filter. Therefore, in principle, the UV mirror is not just as simple as protecting the mirror. In the final analysis, how much the filtered UV light can improve the image quality is a problem. I don't know if it is more effective in places with strong ultraviolet rays on the plateau. Interested readers can try it.