Dark corners in a night photo often look harmless in the original. As soon as you brighten them, however, clear noise often shows up. This is exactly where Marko starts in this tutorial. He works with a photo of a church that contains many black areas, and shows you how to remove slight noise in the shadows without making the image look flat or muddy.
The video is only loaded from YouTube when you play it – data (including your IP address) is then transmitted to YouTube/Google. More in the privacy policy
Two approaches are used here, which you can combine well: first, the targeted removal of color noise using the color sliders, and second, the careful dosing of luminance noise and the darkening of the shadows with a mask. You can easily follow the steps in your own development software for RAW files.
Key Takeaways
- Noise occurs mainly in dark areas of the image and becomes especially visible when brightening.
- Color noise can usually be reduced generously without damaging the details.
- With luminance noise you need a delicate touch, because too much reduction takes the structure away from the image.
- Darkening the shadows with a mask reliably hides noise, as long as you set the mask cleanly.
- Since you are working with RAW files, you can adjust all settings afterwards at any time.
Step-by-Step Guide
Step 1: Making the Noise in the Shadows Visible
Before you change anything, you should know where the problem lies. First take a look at the histogram. At ISO 800, the image initially still looks usable, but the shadows are heavily bunched at the left edge. As soon as you raise the exposure in the dark areas, the noise clearly stands out. Zoom in on the bottom left and right corners, because that is where the noise gathers most strongly. This check shows you where you will need to start later.
Step 2: The Basic Development Before Noise Removal
Before you turn to the noise, you should establish the base image. First, raise the details in the tree and in the water a little to bring out the structure. Then pull back the highlights, because the highlights in the sky are already very blown out. For testing, you can go all the way to the limit to see how the image behaves. Afterwards, you can bring the highlights back in a little so that the image does not look unnaturally dark. A slight increase in vibrance rounds off the basic look. You can also adjust colors and saturation now, for example toning down the dominant green in the background so that the focus is on the church. It is important that you only tackle noise removal once you like the base image, because every later change to brightness or vibrance amplifies the noise again.
Step 3: Removing Color Noise
With color noise, it is best to be bold. Zoom in close on the image area where color artifacts are visible, and initially drag the color noise slider all the way to the limit. This way you quickly see which areas are affected. In the window, small green-purple artifacts show up, which at the maximum value merge into a uniform, slightly gray-purple tone. Important: the pixels themselves are hardly changed in the process, only their color information is smoothed. You can therefore pull the slider back afterwards. Marko decided on a value of 64 as the basic setting. With color noise, you therefore do not need to be squeamish, unlike with luminance noise, which directly affects the structure of the image.
Step 4: Dosing Luminance Noise with Noise Reduction
Luminance noise is the real challenge, because here the brightness values of the pixels change, and with them the structure of the image. For comparison, you can first set the noise reduction to zero and then drag it all the way to the limit. At maximum reduction, the image looks drawn or painted, which can create an interesting look but costs many details. Then pull the slider back until you have found a balance. In the window, Marko set a small tick; at around 10 to 13 the noise remained tolerable, while the details in the window were preserved. What matters is that a little grain is retained, because a completely smooth image quickly looks artificial.
Step 5: Darkening the Shadows with a Mask
Another approach is not to brighten the dark areas, but to darken them a little more. This way, the noise largely disappears into the black. To do this, use the brush to apply a mask to the upper part of the image, that is, the background with its gray-brownish mishmash. Adjust the size of the brush before you paint over it. Instead of lowering the exposure, in this case you pull down the blacks, for example to minus 30. This value initially looks very strong, but since you are working with a RAW file, you can correct it at any time and adjust it until it suits your image.
Step 6: Readjusting the Mask
After applying it, you should check the mask. With a brushstroke like this, you often also catch areas that shouldn't be affected, such as part of the roof. That's not a problem, because you can refine the mask with the brush in erase mode. To do this, select the mask point, go to erase, and remove the excess area. When touching up, pay particular attention to edges where the sky and buildings meet, since transitions look messy there most quickly.
Step 7: Finding the Balance and Evaluating the Result
Finally, you look at the image as a whole. The noise should have decreased significantly, but not disappeared completely. This balance is the real core of post-processing. If a spot still has too much noise, you can readjust it in a targeted way without repeating the entire development process. According to Marko, removing noise is no great witchcraft. It is much more important that you find the balance between calmness in the image and preserved detail.
Summary – Removing Noise in Night Photos: Specifically Reducing Shadow Noise and Color Noise
You have seen that noise in the shadows of a night photo can be tamed with a few targeted steps. Start with the basic development, then reduce the color noise generously and dose the luminance noise with care. In addition, you can darken the shadows using a mask and refine the mask if needed. Always keep in mind that an image with some grain often looks more natural than one that has been completely smoothed out.
FAQ
Why does noise appear so strongly in the dark areas of the image in particular?Dark areas contain little light information. When brightening, sensor noise therefore becomes much more visible.
What is the difference between color noise and luminance noise?Color noise appears as colored artifacts such as green or purple pixels. Luminance noise, on the other hand, appears as a grainy brightness structure in the image.
Why shouldn't I set the noise reduction too high?Too much reduction smooths the image so much that it looks painted and loses a lot of detail.
Can I still adjust the settings later?Yes, with RAW files all settings can be changed at any time afterwards, including the mask and the black tones.
