Those who follow my blog and articles might have read the first part of the post-processing mistakes module with the subject of halo. This time I would like to talk about another issue called colour banding.
When we are working on our photos either in Photoshop or Lightroom, there could be circumstances when we might come across with some editing issues, and the most noticeable of them is colour banding or in other words, posterizing.
For a better understanding, I think the best way is to demonstrate how does banding look like. So let’s compare the two images above! Are we able to spot the differences? I believe that it’s quite noticeable that the left picture just looks fine but the one on the right side seems relatively unrealistic with such weird lines in the sky. This phenomenon is either called colour banding or posterization. To know what it is, what causes it and how is it possible to avoid it, just stay with me and let’s look into it together!
What colour banding is?
First, let’s imagine a picture with pretty much the same colour in a large area- like a blue sky or sunset. In some situations, those backgrounds have different colour shades with lighter and darker tones. If the transitions between tones are smooth and continuous, we won’t be able to recognize the colour change.
However, when there are insufficient colours accessible for smooth transitions, especially in dimmer tones, the result is an ugly picture with lines on it, which is called colour banding.
If banding appears on our images it means that we might pushed our post-processing too far. Although, if we have areas of the picture where there are many objects, banding will be harder to spot. But despite this, I really don’t recommend ignoring it and hoping that others won’t notice the problem!
How to prevent banding on our images?
- First and foremost it’s really important to shoot in RAW as I said before because we don’t have much chance to fix banding with the in-camera compressed JPEG format.
- When we are working on RAW files in post-processing, be careful to not push the values too far, otherwise, banding might appear on the images. In that case, the best what we can do, start the editing from the beginning and change the values. To apply modifications in Camera Raw also decreases the chance of the occurrence of banding.
- On the other hand, we can try to change the Bit depth to 16-Bit instead of working with 8-Bit. The greater bit depth means that we have more colours accessible. Unfortunately, JPEG’s are only 8-Bit files, and we can’t change their Bit depth, however, it’s very easy with RAW files.
How to recognize banding?
– The most apparent way to identify banding is visually observing an image. Although, there could be circumstances when colour banding is not obvious and hard to see, but if we take a better look at the image we might recognize it then.
– If we can’t see any banding with our naked eyes the other way to identify it, just simply have a look at the histogram. In case we discover gaps in the histogram, it means that we applied too many changes when adjusting levels or curves and might have banding in our image.
Let’s compare the histograms below!
The histogram on the left side looks “normal”, while the one on the right side has many spikes and looks strange. To make things more clear, the first image is the unprocessed RAW file, and the second one is after we applied too many changes on it. The difference is visible. Those spikes on the second picture are breaks in the tonal range.
– Alternatively, we can use an easy trick to check banding on our images with Solar Curve. All we need to do is just create a new curves adjustment layer, then place 6 different points to the curve with the same distance from each other and drag them to make a wave shaped curve like that:
How to fix banding?
Now that we have found banding on our image we can think about how to remove it. My personal opinion is don’t remove it, start the editing from the beginning and change values carefully, then check the results with one of the methods shown above.
However, there are different solutions for fixing the banding issue, like adding some noise to the image or increasing the bit depth of the picture, but I believe these techniques are not perfect. They remove some of it, but not all of it and banding will only be less visible than it was before. So if I want to be satisfied, then I have to start the post-processing all over again.