Getting around the “De-ghosting” Problem

by | Jan 13, 2023 | Tips & Tricks | 2 comments

Everyone loves sunset photos! But photographing a sunset is not easy. The dynamic range between the bright sky and the dark foreground is too incredible for the camera to pick up. If I expose for the sky, the result may be nice color, but at the expense of a dark foreground. If I expose for the foreground, the sky washes out.

There are many ways to work around this problem while in the field, like embracing the light and airy feel, or using filters.

But there are also many ways to work around the problem in your photo editing software later. One of these ways is to photograph a middle-of-the-road exposure on-location, and then push your photo editing program to its limits, retrieving as much information in the sky and foreground as possible. With modern camera technology, this is a pretty good option. In fact, it’s the way I’ve been approaching high-contrast situations recently. Problem is, I often get middle-of-the-road results.

I ran into this issue when photographing a gorgeous sunset in Connecticut over the holidays. Lydia and Sarah were shooting the Joyful Pictures STS Assignment and in the meantime, I crouched low over a jetty of rocks, angling my camera to capture the low-lying sun in a ribbon of vibrant color. How to walk away with the tranquil beauty before me?

Here is a photo exposed in-between. Not the greatest, right? Some of the sky is washed out, and the foreground is super dark, but it is photographed in RAW and I could just hope for the best in editing on my computer.

1 middle of the road 221230_James Staddon_8612 Canon EOS 5D Mark III, 16 mm, 1-250 sec at f - 4.0, ISO 100

And here’s how it looks after editing.

221230_James Staddon_8612 Canon EOS 5D Mark III, 16 mm, 1-250 sec at f - 4.0, ISO 100

Not too bad. I could certainly use this. But there are some problems with it.

First, the sky is still kinda bright, especially around the sun. There isn’t the gorgeous sunset tones I remember there being in real life.

Second, there is a lot of noise visible in the rocks when you look at the photo at 100%. Some noise is fine, but this kind of noise is very noticeable, with banding and multi-colored pixels. I have pushed the editing too far. There was not enough information in those extremely dark areas of the photo.

2 grain in rocks at 100

I could solve the noise problem by returning the darkest parts of the photo into darkness again and at the same time lose my rocks into a shapeless mass of black.

Thankfully, since pushing the photo editing program to its limits is not the only way to solve the dynamic range problem, I decided to try another approach.

While in the field, knowing from experience that the middle-of-the-road approach probably wasn’t going to be ideal, I had photographed three different exposures of the same scene.

3 three different exposures

With these three photos, I have the ability now to work around the high contrast problem using photo blending and HDR editing techniques.

At first, merged together with Lightroom’s Auto HDR function, they don’t look any better. Here’s the unprocessed:

4 merged HDR 221230_James Staddon_2 Canon EOS 5D Mark III, 16 mm, 1-15 sec at f - 16, ISO 100

But the flexibility in editing three merged photos is so much better! Here is the processed version:

221230_James Staddon_2 Canon EOS 5D Mark III, 16 mm, 1-15 sec at f - 16, ISO 100

Looks amazing! Or so I thought. But take a closer look at the shadow areas.

5 100 on hdr

Some of the rocks, on the left side of the photo look great! But then there’s this abrupt line, after which all the dark areas to the right are awfully grainy!

HDR merging was supposed to fix the grain in the shadow areas. But something went awry with the auto merging.

In the past because of this, I have shied away from using Lightroom’s HDR merge feature. It just doesn’t get it right often enough to make me want to use their tool.

However, this week I took the time to look into it. And I think I found the source of the problem!

While Lightroom’s HDR merging feature doesn’t offer many options, one option it does give me is “De-ghosting”. When “Medium” de-ghosting is selected, notice the area affected by the overlay.

Right along the awful line of grain!

6 overlay medium

I ran some tests. When “High” de-ghosting was selected, the overlay looked like this:

7 overlay high

And you can guess where the grain began! The grain even extended up into the sky:

8 High 100

Whatever “de-ghosting” was, it was ruining my HDR merging. So, I simply selected “None”. And this got rid of the all the grain! The HDR merging was actually working properly:

221230_James Staddon_3 Canon EOS 5D Mark III, 16 mm, 1-15 sec at f - 16, ISO 100

But now I spotted a new problem. You’ll notice that in the bottom right corner of the photo, the beautiful wave had disappeared. When I selected “None”, it took away the grain but merged the photos in such a way that it also took away the wave.

So, I could just say this was fine. But I didn’t feel like giving up. I liked the wave too much!

So I decided to take things one step further. In Photoshop, I opened both the “Medium” and “None” images and merged them together using a layer mask. In the “None” photo, I kept the nice looking rocks. In the “Medium” photo, I kept the nice looking wave.


And with them both merged together, I got exactly what I wanted! Gorgeous, rich color in the sky, smooth, wave-swept rocks in the foreground, and a perfectly shaped wave beckoning the viewer to enjoy the pebbled tranquility.

221230_James Staddon_2 Canon EOS 5D Mark III, 16 mm, 1-15 sec at f - 16, ISO 100-2

Pebbled Tranquility
Point Beach, Milford, Connecticut
Download as a free desktop background in the next Latest from Lenspiration update (subscribe)

Get each article as soon as it goes live!

Recommended Ebook


  1. Brieanna

    Oh, my. I hope I can edit like that some day! Will you be doing a course on post-processing sometime soon?

  2. James Staddon

    Yes! I’m working on the Composition one right now, but Editing is the next one in the queue!


Submit a Comment

Your email address will not be published. Required fields are marked *

Send the next blog post straight to your email inbox!

Thank you for subscribing!