Fundamentals of audio and video (5): hue, brightness and saturation I in CIE color space

When writing an image, we know that pixels make up a picture. What is that pixel? Simply put, it is a color point. What color is it? Or how to measure it? How are the colors we perceive with our eyes represented by data?

The next few articles will be introduced according to this idea.

When writing an image (1), we know that color is a result of our visual system's perception of visible light, and the frequency of the perceived color light wave is certain.

Perceptual result is a particularly vague word, so CIE (remember this word, it will often appear later) defines these feelings, and uses three characteristics of color to distinguish colors, namely hue, saturation and lightness. These are the distinctive features inherent in color.

Because these three characteristics of color are formulated according to subjective perception results, it should be easier to understand.

Look at the picture above, the hue is represented by a circle on the color ring, and the colors on the circle have the same saturation and lightness. Look at the picture on the right. Is the easiest attribute to distinguish colors. It represents the visual system's perception of the colors presented in a certain area. It can be described by words such as red, orange, yellow, green, blue, blue and purple. In technical terms, the perception of hue is called color. Such as light blue and dark blue. Black, white and gray indicate no color. As we often say, a colorless life.

As shown in the figure, the colors along the radius have the same hue and lightness, but their saturation is different. Saturation is used to distinguish the brightness of colors. When a color is mixed with other light components, it is considered to be less saturated. A fully saturated color is a color without white light.

The seven colors shown in the example have the same hue and lightness, but their saturation is different. The saturation on the left is the shallowest and the saturation on the right is the deepest.

Brightness is the perceptual attribute of visual system to how much light is radiated or emitted by visible objects. For example, a lighted candle looks brighter than an incandescent lamp. This word is more emotional, but it can be understood by comparing color and saturation. Tone can actually be understood as different colors we often say, while saturation can be understood as the intensity of colors.

Its value is 0- 10. Among the seven colors on the right, they have the same hue and saturation, but different lightness. The lightness at the bottom is the smallest, and the lightness at the top is the largest.

Since the subjective perception value of lightness cannot be measured by physical equipment, CIE later defined brightness, indicating the intensity of light reflected or emitted per unit area. This is a measurable representation, and understanding it is enough.

The commentary said that the saturation and lightness of colors are still unclear after reading the article. Well, I thought deeply afterwards, which is really important, because the distinction between lightness and saturation is also the key to understand the development of the next color space and color measurement method.

So next, I want to clearly distinguish these two attributes of color in vernacular Chinese. First of all, we should understand that CIE's hue, saturation and lightness are completely independent and indispensable when describing a color. Why? Next thing you know.

If you want to describe lightness in one sentence, it is darkness and dawn ~

Well, speaking of which, you may suddenly realize. If you still don't understand, please read on.

If the sun shines on the blue sky and white clouds, please look in the direction of the sun, and you will find that the sky next to the sun is white. Then gradually look away from the sun and you will find that the sky is getting bluer and bluer. If you have a pair of clairvoyance eyes, you can see the other side of the earth, then you will find that the sky slowly turns gray, darkens, and then turns black where the sun can't reach, that is, below the horizon.

This process is to change the color when the lightness, hue and saturation are unchanged.

Having said that, we may understand that light is light. We can't see anything without light.

But we may not care too much in life. Light not only affects what we see and what we see.

As mentioned above, from dull to extremely bright, the color changes are as follows: black->; Diablo->; Darkness->; Dark primary colors->; Primary colors-> Light primary colors-> white

As shown above:

So in the color model, we usually describe the change of lightness as 0 to 10, that is, from black to gray to white. When the lightness is 0, that is, black, any color is black. When the lightness is 10, which is white, any color is white.

If you can use PS, you can create a new layer, fill it with red, and then move the brightness from black to white to see the color change. As shown in the figure below:

After talking about lightness, it should be easier to understand saturation. In layman's terms, saturation can be called purity, or brightness of color. Its representation of radius is the process from saturation 0% to saturation 100%.

When the saturation is 0, the colors we see become lightness values, that is, black, white and gray.

Therefore, when the fixed lightness is a certain value (it cannot be 0 or 10 to see the color clearly), the process of saturation changing from 0% to 100% is intuitive, that is, from: the gray level of a certain value->; Grey original color-> Lighting primary colors->; Pure primary color

Namely 1.2 section example diagram:

Frankly speaking, it is pure red, slowly adding a certain proportion of gray, and finally becoming gray.

The fully saturated color in the textbook refers to the color without white light. This sentence is also true, so how to understand it?

First of all, explain from the definition of color, because color is our perception of visible light, and visible light belongs to electromagnetic waves. If the color we see consists of only one wavelength of electromagnetic waves, it is completely saturated.

To put it more bluntly, it is a certain proportion of gray that we slowly add, which is actually a certain proportion of white light.

If you open the color picker on PS, you can see the following picture.

The leftmost part here, from bottom to top, is the change of lightness. From right to left, it is the change of saturation. It can be seen that at the bottom of the figure, when the lightness value is relatively low, the color change caused by the change of saturation is very small.

To understand this, we must first understand the application of color. In fact, you can understand when you think about it. In which scenes do we use colors? If it is divided into two categories, one is display, and the other is paper operation such as printing or drawing.

Needless to say, the display can show excellent properties such as hue, saturation and lightness, while our display can generally adjust brightness (lightness), saturation and even contrast, so it can simulate the real scene well. On paper products, converting the brightness from black to white into pigment, and then watching it under the light, can also well simulate the relationship between light and shade of the real scene.

Having said that, why did you jump to the color space again? To solve this problem, we must first find out why CIE defines these properties of color, just as we define Cartesian coordinate system in mathematics. In the two-dimensional coordinate or three-dimensional space, we define the coordinate system in order to accurately define or find out a certain point, just like giving every household a house number.

The properties of hue, saturation and lightness are also to find out which color it is in the color space.

But one thing, color space, due to the development of history and science, has produced many color spaces to describe colors.

They are synonyms for each other. To put it bluntly, it is a thing, but it is called differently on different occasions.

Color space is mostly used in mathematical scenes, usually represented by a three-dimensional model, that is, three-dimensional coordinates representing three parameters are specified to describe the position of color in the color space. And these three parameters, there is no straightforward explanation of what color it is, it depends on its coordinates. It means you have to take the coordinates and look in the color space to know what color it is.

For example, the color space (hue, saturation and lightness/value) of HSB/HSV directly uses hue, saturation and lightness as coordinate axes. Tone is calibrated by angle, red is 0 degrees, cyan is 180 degrees, saturation is expressed by radius, and lightness is expressed by vertical axis. This color space should be familiar by now.

Because the visual effect of color is based on equipment and materials. When the equipment and materials change, the corresponding color effect will also change. Therefore, we call such color spaces device-related color spaces, such as RGB and CMYK color spaces. When different devices with different color rendering modes represent the same color, the corresponding color values are different.

On the other hand, CIE L a b* color space established on the basis of HSB (discussed later) directly uses a set of numerical values to simulate human color perception, instead of using a set of required numerical values to drive specific devices to generate colors, and its color values are independent of devices, so this color space is called device-independent color space.