# 2 The photo should be properly exposed Sure, but what does that mean? And what is it about the animal, the exposition ? Exactly the exposure is the time how long is the film or sensor on the light. In practice, it is inaccurately used for an exposure value (EV) , which is the result of how long the illumination takes place, what is the sensitivity set, and what is the aperture number (N). EV = 2log2N - log2t Brrrr, Fuuu, what is it, math ?! Yes, but it's not as bad as it looks at first glance. This ugly pattern tells us that the photo is the lighter the lower the aperture number and the darker the shorter the time . This formula applies to ISO 100. Fortunately, ISO is linear, so if you double ISO, EV will double. Note: The exposure is measured at a negative logarithmic scale and the lower number means a lighter photo. T he physicists do not complicate us at all, do we? ISO 2x higher ISO means 2x higher exposure Time 2x longer time means 2x higher exposure F number Moving 2 aperture numbers means 2x differentexposure What does it mean by shifting the F number? The aperture number is the ratio of lens size to focal length. And because it works with the diameter of the lens, it is an area and you have to jump in the powers of two. If, for example, we take the aperture of 1.8 as the basis, the shift by one value is about 2.5. It is always multiplied by √2 (= 1.4). In practice, the following series of basic values are used: 1.0, 1.4, 2.0, 2.8, 4.0, 5.6, 8, 11, 16, 22, 32, 45. Example of time dependency on aperture at constant ISO: F number 1 1.4 2 2.8 4 5.6 8 11 16 22 Time [s] 1/500 1/250 1/125 1/60 1/30 1/15 1/8 1/4 1/2 1 Example of time dependency on ISO standing curtain: ISO 50 100 200 400 800 1600 3200 6400 12800 25600 Time [s] 1 1/2 1/4 1/8 1/15 1/30 1/60 1/125 1/250 1/500 Okay, but what can we do? The good news is that today's cameras have a very advanced auto and exposure values set for you. But it does not always work perfectly. How does the automation evaluate the exposure? The camera detects brightnesses in the scene. It can take the whole picture into consideration or take into account only the location you are focusing on, depending on the settings and application options. The Samsung camera emphasizes the focus area because it assumes that the main object is there. Then it tries to set the values so that the histogram looks good in the area. Histogram Ajaj, another concept I have not explained yet? Well: The histogram is a chart that gives us insight into what the photo exposition looks like. There is a brightness on the x-axis, the y-axis is the representation of the given brightness in the image. When the value is high, there are many points of just that brightness. The histogram has one or more vertices. The top on the left indicates that there are lots of dark pixels on the photo, the right part marks bright pixels. The ideal histogram has peaks between extreme values. If the left or right steep line is completely at the edge of the histogram, it means the image is underexposed (left) or overexposed (on the right). Properly exposed photo, the histogram only slightly touches the edges A correctly exposed photo, while the histogram strikes the right, but this is due to the range of the sensor. The price for this is the loss of drawing in the clouds. The histogram is pressed to the left and to the right. Yet it is correct because it is a dark scene with light elements. What can you do? Missing darknesses are quite easy to fix in a slightly better graphic editor, but areas without artwork can not be rescued normally (photographers are now catching their heads and shouting at me: RAW!!! and are right, but this is beyond the scope of this article). Here is the photo whis had my daughter photographed. She exposed the landscape in the shadows, so the automaton overexposed the photo. By shifting the black point, the landscape is saved, but the sky is never lost. Overexposed photo After editing in the editor Well, a lot of theory, but how to arrange for the photo to be exposed correctly? I recommend first try the automatic . Today it's pretty good and good scenes are good. The new models even know what you're photographed (called AI - Artificial Intelligence), and when such a mobile knows you're taking a night photo, it does not use a dumb automatic that overexposed the photo, but adjusts the settings to keep the night scene atmosphere. If the automatic fails, we can help her. Even in automatic mode, exposure compensation is available. This is especially necessary if we are shooting a scene with a large range of brightness. Maybe a Christmas tree in the dark, but with lit lights. Automation sets values so that it looks like it's in the day, but we do not want it. Then we have two choices: We click on the screen where we want to evaluate the exposure (this is where you focus) We use exposure compensation and download it in this case down. Exposure compensation What are the typical scenes where the automaton fails? (this may not apply to artificial intelligence and scene recognition) Night The camera over-exposes, the exposure compensation to minus Snow The camera is underexposed, it is necessary to move the exposure correction to the plus Scene with a large dynamic range The camera does not have the full range of brightness, it is necessary to decide whether we prefer drawing in bright or dark areas. It is usually best to sacrifice dark places. Black cat in the snow If we focus on a cat, the cat will be gray and snow like a large white area. Move the exposure to minus. If we focus on a pile of snow next to a cat, the cat will be black and the snow will be gray, you need to shift the exposure to the plus. White cat on snow Probably there will be a gray photo on the gray snow on the photo, you have to shift the exposure to the plus. Black cat on a pile of black coal The photo is probably going to be a gray cat on a pile of gray stones, you need to shift the exposure to minus. Why the automaton fails in large-scale scenes is quite understandable, but why does not the white cat fail on the snow? The automaton assumes the ideal histogram, which has most brightness somewhere in the middle. And if the scene is too bright, shifts them a little to darker, gray. The same thing with black cats in the tunnel, the automatic moves the scene to lighter. That's fine if we want to capture the details, but the atmosphere of the scene goes into the grove. That's why I do not like HDR . This is the process of automatically adapting dark and bright places so that they do not lose the artwork. However, the atmosphere of the scene is lost, especially at night photos that look like a daytime photograph. Why are there any spills, and what is it that spills? Imagine the sensor cell as bucket and light (photons) as drops of water. The photons hit the sensor, causing the electric charge, as well as the water raining into the bucket and filling it. If we let the photons on the sensor or water in the bucket go too long, the maximum charge is reached for which the sensor can no longer go - and the bucket overflows. Whatever falls into it, it can no longer be measured, we would have to use a larger bucket. (The water level is the equivalent of the charge and their value corresponds to the brightness for the given point of the image). When shooting with time and aperture, we ensure that the light comes to the sensor. If there is little, the point will still be scored as black. When the power, as well as the points to be bright, will be evaluated as maximum brightness and the drawing will be lost in such areas. It is even more complicated by the fact that the red, blue and green ones are evaluated separately, so you can lose the drawing on the flower, but the leaves will still be fine. This is due to the different sensitivity of the sensor to different colors. Mobile photo: # 1 SharpnessPrevious part is here:
... View more
