Shulou(Shulou.com)11/24 Report--

When taking photos, we often encounter noise, especially in the low light environment, using a higher ISO, there will be a lot of noise in the photos, which requires noise reduction.

Noise reduction often meets a problem:

If the noise is reduced too much, the noise disappears and the details of the photo are lost.

If the noise reduction is too light, the photo details are preserved, but there is still a lot of noise.

Stack noise reduction can not only reduce noise, but also improve photo details.

1. Stack noise reduction ① first needs to take multiple photos (about 4-9 photos, less effect is not obvious, more effect will not be greatly improved).

The position when shooting can not be changed, it is recommended to put the camera on the tripod. If the camera shakes, now hold down the Ctrl key in ps and click to select all the layers, and then use "Edit-automatically align layers" to align the photos.

② opens Photoshop and selects "File-script-load File on the Stack". Click "Browse" to open the set of photos that need to be stacked. All the photos will be in one document, and each photo will have a layer. (if it shakes, it needs to be aligned according to the method of step ①)

③ hold down the Ctrl key and click to select all layers, right mouse button, convert to smart objects. At this time, all the photos are organized into a smart layer.

④ selects layer-Smart object-Stack Mode. The effect of each mode is different, to reduce the noise of the photo, you should choose the middle value or the average value.

⑤ next, let's enlarge the photo and take a look at the original image and the average noise reduction. (the screen of the phone is small, and the contrast may not be obvious)

We continue to enlarge the photo, the first picture is the original picture, the second picture is the average noise reduction. It is obvious that the noise has been greatly reduced, and after the noise has disappeared, the details of the photo have also been improved.

Continue to zoom in and compare it with the motion picture. Now do you feel the charm of stack noise reduction?

2. Stack noise reduction principle beginners may feel very strange and confused when they hear the word "stack". What exactly does stack mean, and why is it called stack?

Stack is to stack multiple photos together and generate a new photo according to certain rules.

(multiple photos → superimposed → setting rules → generation)

So why can the stack reduce noise?

This starts with the characteristics of the noise. The noise is randomly generated and will appear anywhere in the photo.

When you take the first picture, there is a noise in this place. If you take the second shot, the noise may not necessarily appear in the same position. Continue to shoot photos 3, 4, 5, 6, the same is true.

Average stack ① if you take four photos, one photo has noise in the middle and the other three have no noise. Photos in rgb mode, 0 for black and 255 for white. 0-255 fades from black to white. The color value of the noise is now 100. There is no noise, the color is 255.

② uses the average stack to calculate the average of four photos, as well as primary school math. The final result is about 216. This value is very close to the noise-free color (255).

Median stack ① if you take four pictures in front of the white paper, the middle positions are white, gray, gray, and gray respectively.

② now uses the median stack, which is to calculate the median of four numbers, even with simple mathematical knowledge. The final result is 125.

So, should you use the median or the average stack?

In practical use, the noise reduction effect of the average value is similar to that of the intermediate value, which can be used. You can also try both, compare carefully and make the best choice.

The median also has the function of eliminating sundries. For example, you took six photos in a row, one of which showed a passer-by. If you use the average, there will be remnants of passers-by. Using the median value, passers-by will be removed directly.

Even if you don't need an average, you can easily remove passersby by relying solely on masks.

Took a lot of photos, as far as possible to ensure that "every position captured no one and no car", using the median stack, coupled with mask repair, removed vehicles and pedestrians, leaving only one person as the subject. The main body is the point of interest of this photo, if there is no point of interest, the photo will become boring.

3. The stack needs to take a few photos. Finally, let's talk about an issue that people are more concerned about: how many photos the stack needs to take.

Take the average noise reduction as an example, with the increase of the number of photos, the change of noise reduction will gradually slow down. 4 compared to 2, the gap is very big, but 30 compared to 15, the gap is already very small.

It is recommended to take 4-9 shots, which is enough.

This article is from the official account of Wechat: photography Research Society (ID:shoujiPhotography), by Zhang Xiaozhang.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.