They have in their arsenal many different editing tools, the skillful use of which will help create a true photographic masterpiece. But among them there are several important ones, without which almost no photo editing can be done. Today we’ll talk about one of these tools - Levels. Each photographer develops his own image editing algorithm over time. For me, the first thing I do when I open a photo in Photoshop or Lightroom is look at the exposure. At this point, if the exposure needs fixing, I use the Shadow/Highlight tool ( Shadow/Highlights), and then I start adjusting the Levels ( Levels).
What is the Levels tool?
This tool allows you to adjust the tonal range of an image, as well as its color balance. Thus, using the Levels tool allows you not only to change the exposure, but also to adjust the colors. Of course, there are other tools in Photoshop that can cope with these tasks, but it is through Levels that this is done very simply and quickly.
This tool uses a histogram to give you a visual representation of the tonal range of the image you're editing. Theoretical and practical analysis of the histogram deserves a separate article, but the main thing you need to know about it is that there is no right or wrong histogram pattern. In the Levels tool, the histogram is represented as a graph with a numeric range from 0 to 255, where 0 is black and 255 is white. Thus, the pixels located at the extreme points are completely black or white, not carrying any additional information for detail. Speaking of the histogram in general, if it's skewed to the left, it means your image is dominated by dark pixels and may be underexposed. Accordingly, a significant shift of the histogram to the right may indicate an overexposed image. The middle slider adjusts the midtones or, more simply, the overall gamma. All pixels that do not correspond to the extreme values of the range 0-255 fall into this category.
How do Levels work in Photoshop?
Often, after opening the Levels tool, there is an instinctive desire to brighten a photo using the middle slider. This may produce the desired result, but I would suggest first looking closely at the image for a dominant shade. Often, if there is such a predominant color or shade, it is undesirable. For example, in a wedding photo taken in cloudy weather, the bride's white dress may have a blue tint that you would like to get rid of. At the same time, the warm orange tint in which the entire scene in the photo of a summer sunset is bathed carries an atmosphere and you hardly want to get rid of it. The easiest way to find similar color shades is in the light areas of the photo and, especially, those that should be white. Depending on the lighting at the time of shooting, the hue may be green, purple, blue, yellow, orange, or anything in between.
How to use Levels in Photoshop?
You can apply this tool to any image that needs contrast or color balance adjustments. For example, like this shot of a stormtrooper from the Star Wars cinematic universe. So:
1
. Open the desired image in Photoshop.
2
. Click the new adjustment layer icon at the bottom of the Layers panel and create a Levels layer, or click on the Levels icon in the Adjustments panel, which is located immediately above the Layers panel.
Step 1 - If you need to do color correction
If your image has an unwanted color cast, follow these steps:
1
. In the list of working channels (RGB by default), select each channel (Red, Green, Blue) individually and move the black and white sliders to those parts of the histogram where it begins to move up. If there are no similar shades in the image, the histogram on all channels will touch the edges. Since the image we were looking at had an unwanted tint, this is what the histogram looked like in the green channel. In the following screenshot, areas of the histogram that did not carry any information about color are highlighted in red. Moving the black and white sliders according to the principle described above leveled out the unwanted tint.
2
. As you make this adjustment, you will notice how the image will take on very strong color tones of the channel you are working with at the moment. Don't be alarmed, the colors will normalize once you do this on all channels.
3
. After correcting the color on each of the three channels, you can begin adjusting exposure and contrast.
Step 2 - Adjusting Exposure and Contrast
As we said, Levels also makes it easy and quick to adjust exposure and contrast. In other words, you can use this tool to make highlights, shadows, and midtones lighter or darker. Here's how it's done:
1
. Move the white slider from the right edge to the beginning of the growth of the histogram on the RGB channel. We do the same with the black slider on the left. But be careful not to push the light areas into overexposure, and not to let the shadows fall into darkness. To control the "dropping out" pixels, hold down the ALT key while moving the black and white sliders.
2
. The last step in quickly adjusting the levels is to adjust the middle slider on the same RGB channel. This adjusts the contrast of the image.
- As with any other tool, when using Levels, do not forget about a sense of proportion.
- Don't rush to make drastic changes. Start small, abstract, look at the result and only then enhance certain effects if you think it is really necessary.
- Use the ALT key to check if you're losing too much detail in the highlights and shadows by adjusting the histogram.
- Add a little contrast using Levels, this will make the colors brighter and the drawing more interesting and lively.
Generally Levels in Photoshop is a powerful editing tool, and its ease of use allows you to use it in the process of processing almost every photo without spending a lot of time. A little practice and you will learn to intuitively feel this tool and see where and how it can be used in a particular photo.
The right edge of the histogram scale is the point of maximum brightness (white). If my histogram doesn't reach the right edge, then there are no white spots in my photo. I take the right slider (white) and move it to the left. This tells Photoshop that the brightness at which I stopped the engine should now be taken as white, and all other brightness levels should be recalculated logarithmically.
Those. Initially the slider was set to 255, and I moved it to 200. Now all the points in my photo that had a brightness of 200 (if any) will have a brightness of 255, all that were 199 will be somewhere around 253, etc. . Those. Photoshop will recalculate all levels, even if I did not have points with any brightness.
Why did they draw a histogram for me under the level line? So that I can see what I'm doing. The rightmost point of my histogram corresponds to the brightest point of my photo. If I move the right (white) slider towards it, the brightest point of my photo will become completely white. And if I continue to move the slider to the left, then even my not brightest tones will turn white. And those that remain to the right of the engine will be even more white, i.e. several tones will turn into one, white (Fig. 5). I'm starting to lose the undertones in the highlights. Those. most often I don’t want to drag the white slider to the left of the right edge of my histogram (see Fig. 6).
The left engine (black) is completely similar. It determines what level my current photo will become pitch black after processing. The rest are recalculated logarithmically. Those. I can push it in until I reach the left edge of my histogram, which will mean that the darkest tone of my photo will become completely black (if I want it, of course). By moving it further to the left edge of the histogram, I begin to lose detail in the shadows, because... all the tones remaining to the left of the slider turn into one tone - black.
The gray slider (middle) determines what tone will be medium gray (128). I can move it to the right or left. By moving it to the right, I make gray those tones that were originally lighter than gray, i.e. I darken the photo. At the same time, I only redistribute the tones, black remains black, white remains white, and all intermediate tones become darker. And when I move it to the left, I make gray those tones that were originally darker than gray, i.e. I lighten the photo.
To the right of the histogram there are three more eyedropper buttons. Pipettes are an alternative way to set the position of the sliders. Those. Instead of moving the right (white) slider to the desired value, I can “take” the right (white) eyedropper, and “show” on the image itself exactly what tone I want to turn into white. This is absolutely the same as shifting the right slider to the brightness of the point I selected, which, in turn, is absolutely the same as entering the brightness value of the point I selected into the right one from the “Input Levels” windows. All three methods - entering a number in the window, moving the slider or pointing with an eyedropper, do the same thing - they show which level should be considered white. Similarly, the middle (gray) eyedropper and middle window correspond to the gray slider, and the left (black) dropper and left window correspond to the black slider.
ACQUAINTANCE
In this lesson we will get acquainted and understand how “Levels” work in Photoshop.
Despite the fact that “Levels” are intuitive and simple to use, they hide great possibilities.
Let's try to figure out which ones exactly.
Traditionally, you can create a “Levels” adjustment layer in Photoshop in several ways:
1. In the “Layers” panel, at the bottom, click on the “Create a new adjustment layer or fill layer” button and select “Levels” from the list.
The creation of an adjustment layer is confirmed by the formation of the “Levels” layer in the Layers palette.
2. In the “Correction” panel, select the “Levels” icon.
3. In the “Layers” menu, on the “New Adjustment Layer” line, select “Levels”.
The difference between this method is that after calling “Levels,” a window appears in which you can set the name of this adjustment layer, use it as a clipping mask, select the layer with a certain color, and what’s most interesting is to immediately apply the blending mode and change the opacity.
When you create an adjustment layer, the Properties panel opens.
If suddenly the panel does not appear, go to the “Window” menu and select “Properties”.
These three methods create a “Levels” ADJUSTMENT LAYER, but there are also other “Levels” that are hidden under the “Image” menu tab, in the “Adjustments” line. They can also be called up by pressing the key combination Ctrl+L.
These are the same “Levels” in their functional purpose with the same adjustments and settings.
The difference is that Levels, brought up from the Image menu, affects directly a specific image, while a Levels adjustment layer affects all underlying layers in the Layers palette, except when used as a clipping layer. masks. We can also disable or enable the adjustment layer settings at any time during the processing process by clicking on the button in the form of an eye to the left of the “Levels” icon.
Also, at any time we can make changes to the “Levels” adjustment layer, while the “Levels” located in the “Images” tab are applied once and for all. But they have their advantage: they can apply correction not only to the image, but also to channels, both to one of them and to a group.
These “Levels” can also adjust layer masks and change their contrast.
Which “Levels” are better? The question is not clear-cut. Each of them is used for its own purposes and it is important to be able to use both.
Let's now understand the Properties panel.
In the central part of the “Properties” panel of the “Levels” adjustment layer there is a HISTOGRAM.
A histogram is a kind of graph of the distribution of the number of pixels with a certain LEVEL of brightness.
When you open an image, the program analyzes it and recalculates the number of pixels in each brightness level, after which a histogram is built.
That is, in Photoshop there are 256 brightness levels (from 0 to 255), where “0” is the level of the darkest (black) pixels, and “255” is the level of the lightest (white) pixels, and these levels are plotted horizontally. The number of pixels in each level is formed vertically.
When we look at a histogram, we seem to be dividing it into three areas: darks, midtones, and highlights. By analyzing these areas, we conclude for ourselves how many dark, medium or light areas there are in the photo. Next, you can decide what actions need to be taken to correct the tonality, i.e. difference in image brightness, if necessary.
In the upper left corner of the panel, next to the “Levels” icon, there is a “Masks” button. By clicking on it with the left mouse button, we will have access to control the mask, which is created simultaneously with the “Levels” layer.
The panel still has a button for calculating a more accurate histogram, but it does not have any global changes in the appearance of the histogram or in the operation of the adjustment layer.
At the bottom there are general purpose buttons. From left to right: “Cut off the layer from the rest”, i.e. the impact will only apply to the underlying layer, and not to all; “View previous state”; “Restoring default correction values”; “Turn on/off layer visibility” and the “Delete adjustment layer” key.
We'll look at a few simple techniques for correcting color photographs using levels in this short post. The use of these methods will not cause any difficulties in use even for novice Photoshop enthusiasts, and their use gives excellent results.
In the last post, I already wrote about the use of brightness levels, although in relation to a black and white image (read) for color photos, the principles of their use are not much different, although there are some nuances here.
For work, I chose this photo, in principle it is not so bad, but it can be made more attractive and get rid of a large amount of blue color, which is clearly in abundance here.
Open the photo you will be working on and create a “Levels” adjustment layer using the “Adjustments” palette by clicking on the appropriate button.
Using Brightness Level Sets
In the palette that opens under the levels chart, move the markers for dark tones (left) and light tones (right) to the beginning of the histogram and, if necessary, use the neutral color marker (in the middle) to make your photo darker (if you move it to the right) or lighter (by moving the marker to the right ). This is the result I got
However, for such a correction it is quite possible to get by with sets of brightness levels, which are located in the “Set” drop-down list. Click on the button shown in the picture below
Now your photo is back to its original state and you can use the kits. Open the drop-down list and select any of the proposed sets
If you are not satisfied with the result obtained, cancel the set action using the button shown above, or correct the result obtained using the markers under the histogram.
Actually, that’s all the correction except for the fact that you need to get rid of the excess blue color, this can be done by correcting the brightness levels for a separate channel.
Correcting brightness levels in a separate channel
Up to this point we have adjusted brightness levels in the composite RGB channel, in order to work with any individual color composing the image, you need to select a separate color channel. I have a lot of blue in my photo that I need to get rid of. To do this, I'll select the blue channel from the channel list.
I’ll move the neutral color marker a little to the right side by 0.79, and the blue color will become smaller, and green will take its place. And I added a little light tones and moved the right marker to 226. As a result of these actions, I got this photo
As you can see, using brightness levels to correct color photos is quite simple and you can get a very good result. Read about a quick way to correct photos using curves. Using levels to create a color balance layer mask is described in the article
GENNADY MEERGUS
HOW TO WORK WITH LEVELS IN ADOBE PHOTOSHOPBAR CHART
To understand levels, you must first understand the histogram. It looks something like this:Fig.1. bar chart
A brightness histogram is a graph in which the horizontal line shows the brightness, from zero (absolutely dim, black) to a maximum of 255 (absolutely bright, white), and the vertical line shows the number of points with this brightness. Thus:
A. for a completely black image we get a thin, thin column on the left (many black dots, and no others), for a completely white photo - a column on the right, for gray (128) - a column in the middle;
b. for, for example, a gradient - an absolutely smooth transition from black to white, we get an even rectangle on the histogram, because there is the same number of points of each brightness;
V. in most real photographs from real life we will get some kind of curve. If it has a visible hump on the left side of the histogram, it means the photo has more dark tones, low key. If it’s on the right side, it means it’s so light, high key;
d. If the histogram does not reach the left edge, then the photo does not have a completely black tone, but only gray; if it doesn’t reach the right one, it means there is no completely white. If it doesn’t reach either one or the other, it means that the photograph as a whole is gray and sluggish;
d. If the histogram hump looks like it consists of separate “teeth” with breaks between them, then there are few intermediate tones in the photograph, i.e. In general, there are few individual tones, the photograph is contrasting. If the hump is smooth, it means the photo is soft, with smooth transitions. Solarized photographs (with an extremely small number of halftones or no halftones at all) look like two or three or four thin columns in total. Those. there are points of such, such and such brightness, but there are no intermediate ones between them. A completely graphic photograph, with only black and white tones - two columns: the rightmost and the leftmost.
Fig.2. Photograph and its histogram. The hump on the left side is because the picture as a whole is dark, low key. The small hill on the right matches the tones of the white flowers.
COLOR HISTOGRAM
In general, the same as brightness, only it is built separately for brightness in each color channel. Those. there are three of them - red, green and blue. Well, the point is the same: the histogram in the red channel shows how many points there are with the corresponding value in the red channel. If the red histogram is huddled on the left, it means that the photograph in general has few red tones, it is all cyanide, if it is blue, it means that the photograph as a whole is yellowish, etc. If the blue histogram does not reach the right edge, then there is no point in the photo that has a maximum in the blue channel. Likewise, if all three color histograms fall short of either the left or right edges, the photo has few saturated colors. A photograph without saturated colors is what is called “faded” in the language of underexpressed emotions.
Fig.3. A photograph, its brightness histogram and blue channel histogram. The photo has a pronounced yellow tint, so in the blue channel the histogram is significantly shifted to the left.
LOGARITHMIC
Human vision is logarithmic (and so is hearing and other senses). This means that when we see a certain “step” of brightness, then in fact there is multiplication taking place, not addition. For example, a picture of halftones for calibrating monitors, which is published on many sites, is perceived by us as a series of identical steps. We visually perceive this series as arithmetic, but in fact it is logarithmic.
That is, what we see as
A + x = B, B + x = C, C + x = E, etc.
actually is
A * x = B, B * x = C, C * x = E, etc.
The same thing, by the way, is with the notes on the piano, because our hearing is also logarithmic.
LEVELS
When we open “levels” (Ctrl-L) in Photoshop, we see the histogram of our image as it is. Below the histogram there are three sliders. On the right is white, on the left is black, and gray is in the middle.
I can move each of them. What's the point?
Fig.4. Levels
The right edge of the histogram scale is the point of maximum brightness (white). If my histogram doesn't reach the right edge, then there are no white spots in my photo. I take the right slider (white) and move it to the left. This tells Photoshop that the brightness at which I stopped the engine should now be taken as white, and all other brightness levels should be recalculated logarithmically.
Those. Initially the slider was set to 255, and I moved it to 200. Now all the points in my photo that had a brightness of 200 (if any) will have a brightness of 255, all that were 199 will be somewhere around 253, etc. . Those. Photoshop will recalculate all levels, even if I did not have points with any brightness.
Fig.5. Shifting the white slider from 255 to 200 - brightening the image with loss of detail in the highlights
Why did they draw a histogram for me under the level line? So that I can see what I'm doing.
The rightmost point of my histogram corresponds to the brightest point of my photo. If I move the right (white) slider towards it, the brightest point of my photo will become completely white. And if I continue to move the slider to the left, then even my not brightest tones will turn white. And those that remain to the right of the engine will be even more white, i.e. several tones will turn into one, white (Fig. 5). I'm starting to lose the undertones in the highlights. Those. most often I don’t want to drag the white slider to the left of the right edge of my histogram (see Fig. 6).
Fig.6. Shifts the right slider to the right edge of the histogram. Details are not lost, because There are no dots in the image with a brightness greater than 170. The contrast of the image will increase.
The left engine (black) is completely similar. It determines what level my current photo will become pitch black after processing. The rest are recalculated logarithmically. Those. I can push it in until I reach the left edge of my histogram, which will mean that the darkest tone of my photo will become completely black (if I want it, of course). By moving it further to the left edge of the histogram, I begin to lose detail in the shadows, because... all the tones remaining to the left of the slider turn into one tone - black.
The gray slider (middle) determines what tone will be medium gray (128). I can move it to the right or left. By moving it to the right, I make gray those tones that were originally lighter than gray, i.e. I darken the photo. At the same time, I only redistribute the tones, black remains black, white remains white, and all intermediate tones become darker. And when I move it to the left, I make gray those tones that were originally darker than gray, i.e. I lighten the photo.
Fig.7. Shifting the middle slider from 1 to 1.3 - brightens the image without losing details.
To the right of the histogram there are three more eyedropper buttons. Pipettes are an alternative way to set the position of the sliders. Those. Instead of moving the right (white) slider to the desired value, I can “take” the right (white) eyedropper, and “show” on the image itself exactly what tone I want to turn into white. This is absolutely the same as shifting the right slider to the brightness of the point I selected, which, in turn, is absolutely the same as entering the brightness value of the point I selected into the right one from the “Input Levels” windows. All three methods - entering a number in the window, moving the slider or pointing with an eyedropper, do the same thing - they show which level should be considered white. Likewise, the middle (gray) eyedropper and middle window correspond to the gray slider, and the left (black) eyedropper and left window correspond to the black slider.
LEVELS AND CONTRAST
The original photo had a maximum of 256 brightness levels, from 0 to 255. By moving the right and left (white and black) levels, I leave fewer brightnesses, “cut off” those that remained to the right of the slider, they become white. In other words, when working with the right and left sliders, I involuntarily make the photo more contrasty.
What happens to the contrast when I move the middle slider, for example, to the right? To the right of the slider there are fewer points that need to be distributed over half the range, therefore there will be fewer brightness levels there than there were before processing. And on the left there are more points, which also need to be distributed over half the range. It turns out that by darkening the photo (middle slider to the right), I make the highlights more contrasty and the shadows softer. Similarly, by brightening the photo (gray slider to the left), I make the highlights softer and the shadows more contrasting.
Sometimes after such processing you can see “teeth” in the new histogram.
OUTPUT LEVELS
Under the ruler with three sliders there is also a gradient scale and a ruler with two sliders. This is the opposite operation to the one described above. Those. there we said that a certain level that was less than white should be made white. And here we say that the white one needs to be lowered to the level on which I will put the right engine. Accordingly, the left slider shows what the former black will become. After such an operation, there will be no completely white tones left in the photograph; the picture will become grayer, more faded, but also softer, with a large number of intermediate tones.
Fig.8. The output level is shifted from 255 to 218. The contrast is noticeably weakened, the picture has become more sluggish, and the white tones have disappeared.
Output levels are not the opposite of input levels. Those. if I used input levels and made a certain color white, then all tones brighter than it were discarded and became white. With output levels I can turn this white into some kind of gray, but I won’t get the lost tones and transitions back.
But on the contrary - it is possible. By the output levels I can tell that the white is turning to grey. OK, done. Then with the input levels I can say the opposite: take this gray and make it white. This trick will work.
COLOR LEVELS
Color levels allow you to carry out the above operations with each of the three channels separately. This is a good way to correct (or change) the color balance. For example, there is a yellow tint in the photo (as often happens from Kodak minilabs). Most likely, in the histogram of the blue channel, I will see that the blue color does not reach the right edge of the histogram, but is huddled in the left half. By moving the right slider approximately to the edge of the histogram, I will increase the blue value to 255 (the absolute maximum) at those points where it was the maximum of all those available in the photo. If there was a color imbalance, then after the described operation with all channels, the colors will become purer and more neutral, without pronounced deposits. In good cameras, with carefully calculated formulas, using levels you can sometimes even correct erroneous white balance or correct color lighting. And in some - it’s not possible. :-(In my previous camera Sony D770, an error in the white balance was tantamount to throwing away the picture. I had to carry a gray card with me and constantly measure the color of the light before shooting. In the current one - Olympus E-20 - setting the balance white works clearly by eye, and even if there is an error, it can sometimes be corrected. In general, you cannot rely on this to set the white balance carefully. But I digress.
Fig.9. Enhancing the orange flavor by shifting the middle slider in the red channel to the left
AUTO LEVELS
The AutoLevels command (ctrl-shift-L) does the following: in each of the color channels, it moves the right slider to the right border of the histogram, and the left slider to the left border. Those. “cleanses up” the colors in the photo. Most often, this visually improves the photo, especially after a stupid scan by an inexperienced dummie. Colors become visually purer, and the picture, most often, becomes brighter and richer. This is because the photo now uses the entire available range of brightness. Those. the point with maximum red has now become completely red, etc. The point at which all the colors were at their maximum in the image became a monochromatic neutral, probably white, although this is not a fact, because this operation works with each color channel separately, but does not affect the brightness channel.
Fig. 10. Picture before and after operation "Auto-levels". It is clearly visible that the operation has increased the contrast excessively. In this case, the underexposure was intentional, the exposure was determined point by point on the white flower (push process). Finishing this shot, of course, requires more subtle tools than auto-levels.
But, as always, there is a “but”. Firstly, if the photograph had a pronounced coloring, now it will not have it, because Photoshop will “stretch” the histograms of underdeveloped color levels (see Fig. 11). Secondly, if the picture, by definition, did not contain any tones, dark or light, then auto-levels will ruin it. If the photo consisted of a cheerful gray color with white specks, now it will be white specks on a black background. And if these were gray stars on a black background, now they will be white stars. Well, and thirdly, as we have already said, using levels increases contrast, which is not always desirable.
Fig. 11. Auto-levels “killed” the pink sunset color of the photo (the brightest tone of the photo became white).
Another problem with auto-levels is boundary values. Photoshop “does not see” the histogram, it “blindly” determines its edge: it simply “pushes” the level sliders inward until it detects a certain number of tones that have merged. Those. This operation has a certain error and with it some halftones in colors and shadows are inevitably lost. In principle, knowing how he does it, you can always do it manually and by eye more accurately and subtly than automatically. In addition, you can manually preserve the coloring, do not touch a certain channel, or in some channel do not move the engine close to the edge of the histogram, but leave it at some distance, in short, play around (until complete creative satisfaction).
LEVELS IN BLACK AND WHITE PHOTOS
In black and white photographs, and especially in portraits, some caution should be exercised with levels. For example, a person is photographed against the sky. Feel free to grab the level and bring it to the right border of the histogram. The sky became completely white, the picture became brighter and more beautiful. What if the person was against a wall, and the wall turned out to be denser (i.e. darker) than the person’s forehead? The brightest place in the photograph will be the person’s forehead, and we will make it absolutely white. You will end up with such a white “bald patch”, absolutely without details, which photographers tactfully call “flare”. In black and white photographs there is only one channel. Often you see pictures with completely white highlights on the arms, legs, shoulders, bellies, foreheads and cheeks. Halftones that have “gone into highlight” cannot be returned. Here it is most appropriate to “play”, i.e. bring the level not to the very right edge of the histogram, but leave it at some distance, so that the desired forehead (shoulder, arm, hip) remains a recognizable cute gray flesh tone, and does not become white. Same with black. It should be remembered that nothing can be corrected.
Fig. 12. Levels in black and white photography. In this image, when working with levels, you should not move the white slider to the right border of the histogram, because The brightest tone is the highlight on the model's cheek and we don't want it to be completely white.
SCAN LEVELS
Many scanners also have a panel for controlling levels. The meaning, in general, is the same: relying on the preview window, you “tell” the scanner which color to consider white, which black, and which medium gray.
This is very important because... the scanner will recalculate the brightness accordingly. If you don’t tell him this, then he can turn the brightest color in the photo into some kind of faded gray, and all this only because the scanner lid or photo frame (taken by the scanner as a reference point) was brighter than the object that you consider the brightest, and which should be the brightest.
There is an interesting difference: in Photoshop, moving the level sliders inward made the photo more contrasty, because forced the redistribution of fewer levels over a larger range. When scanning, it's the other way around. We say which point is the brightest, which is the darkest, and the remainder is distributed by the scanner into 256 levels. Those. we set a smaller interval of brightness, which should be decomposed into the same 256 semitones. This means that when working with levels when scanning, we make the photograph softer, not more contrasting, and add more halftones to it.
CMYK LEVELS
Until now, in order not to make a reservation every time, I have talked about levels in the additive system (RGB). In the subtractive system, we don't add colors to get white, but we subtract them from white, getting black when all is subtracted. Therefore, in CMYK, all panels, except for brightness, look the other way around. Those. the darkest level is on the right, and the lightest level is on the left. Slide to the left - darken, slide to the right - lighten. Well, and accordingly, the levels are calculated not on an absolute scale, from white 0 to black 255, but on a relative scale, from 0 to 100%. 0% is white, of which nothing is subtracted. 100% is black, from which everything is subtracted, nothing remains. In the brightness panel everything remains as before.
LEVELS AND ARITHMETIC BRIGHTNESS
Arithmetic brightness is a vile function made by stupid computer scientists who understand nothing about the logarithmic nature of our vision. When we take out some crappy program, made by a programmer at home on his knees in his free time from Tetris, and increase the brightness there, the program simply proudly adds some value to the brightness at each point. Did this make the photo brighter? There is another way to say it. If there were white dots in the picture, then by adding a certain number to their brightness, we brought them to 255, i.e. they remained white. If there were some “almost white” points, then they could also go beyond 255, i.e. merged with those who were already white. Those. we lose halftones almost immediately.
And what else? And also, if we had, say, black dots, then after adding a certain number to them, they became not completely black, but dark gray. And most importantly, the photo will not become visually lighter, because the logarithmic difference between the lightest and darkest will not change, but only shift. But noise will appear immediately. Likewise in the other direction. In short, arithmetic brightness always spoils the photo, makes it grayer, throws off the color balance and always loses halftones either in the highlights or in the shadows.
Why is this function in Photoshop? Well, for example, for correction. Conversion to CMYK and back is done using formulas close to k/x+b, i.e. the result of the transformation may turn out to be shifted, and then using arithmetic brightness you can return the image to the desired range, and then “treat” it with levels. And we don’t always work in RGB and CMYK; there are other spaces where brightness has a different meaning.
The main thing is that “arithmetic brightness” is not a tool for lightening and darkening, because Levels are such.
Fig. 13. Arithmetic brightness. Increasing the arithmetic brightness led to the disappearance of details in the highlights (clouds). The overall contrast of the image decreases.
CURVES
Curves are, in general, the same levels, only with greater control over the process. When I talked about engines, I mentioned that we set the level for white, gray and black, and the rest are recalculated logarithmically. In “curves” you can control the law by which brightness is redistributed, i.e. directly set the transfer function between what is and what you want. On the X axis - existing brightnesses (what is), on the Y axis - what they need to be converted into (what you want). Gradients are drawn along both axes, showing the density at each given X and Y. The shape of the curve can be changed by placing and “dragging” inflection points along it. Photoshop doesn’t even draw a histogram on the background (some other programs do), because this operation is so surgical that it is almost always necessary to monitor on a real screen what is happening there. This operation is often used in the printing industry to fine-tune the tones output by the printer or when scanning.
Curves are a more subtle and sensitive tool than levels, and their discussion is beyond the scope of this presentation.
