Before we get started, just a few words about High Dynamic Range (HDR) editing. There are a lot of different ways to apply HDR processing to an image. All of them have their pros and cons. The way I use HDR in this post is the most basic use, and I'm sure that the more advanced of you think it isn't the right way to do it. But for you who don't know what HDR is, and why it's useful, I hope you'll find the following a useful stepping stone into expanding your toolbox. You can use this technique with any camera (even the one in your phone!), and it is especially useful on older and/or smaller cameras that don't have a very good Dynamic Range to begin with.
So, what is HDR and why should you care? Simply put, your camera is not as good as your eye at recording big ranges of light versus dark. If you stand in your back yard in the summer, you will note that your eye can see the blue in the sky, as well as the dark under the trees. Similarly, during e.g. sunset, your eyes and brain have no difficulty in seeing the last rays of the sun in the sky, as well as the dark ground in front of you.
For the camera, this poses a problem. Even though camera sensors have come a long way in recent years, they still can't compete with your own eyes when it comes to dynamic range (the range of brightness from black to white). Modern cameras can detect up to 15 stops of dynamic range at most, while your eyes are in the range of 20 or so (combined with your brain that range increases by a significant amount). If you are not familiar with the term 'stop', you might want to google that first, as it is a very fundamental measurement in how cameras tell you how bright something is. For now, I'll just tell you that each stop is a doubling of the amount of light. So, increasing the stop by one doubles the brightness, increasing it with two doubles the brightness twice, etc. A difference of 10 stops is actually a brightness difference 1024 times to the original.
So, cameras can capture highlights that are around 4-8000 times brighter than the shadows, in theory at least. In practice, this range is smaller; the shadow range of that number comes with a lot of visible noise that you typically wouldn't want in your image. Of course, it doesn't matter what the numbers actually are, we are interested in the result. So, let's take a look.
The image above is a scene from the Finnish archipelago, during a nice sunset. It is processed to convey the feeling of serenity; but it is also processed to maximize dynamic range. When I took the image, I realized that the dynamic range of my camera wasn't going to cut it. The sky was much too bright, and the ground was much too dark in comparison. It's not what my eyes saw, but unfortunately my eyes don't take the picture. Here is the raw output of my camera with no processing:
The camera chose to expose more of the ground in this particular instance. As you can see, the ground is visible but still dark, and the sky is almost white. There is another problem as well; the sky doesn't only look white, it is white. Going to the 'Develop' Module in Lightroom and clicking on the highlight warning triangle in the histogram confirms this:
The area marked with red is pure white; that are was simply too bright for the camera to capture. The data in that area is gone; no colors, no contrast, no anything; it is just white. Because of this, reducing the whites or highlights doesn't work. The area isn't white anymore, but is still just a blob of the same color.
That does not look right. The sun and clouds around it are a solid block of color, no detail what so ever.
Typically, you can fix this by deliberately under-exposing your image by a stop or two. You can do this either in Manual mode or with Exposure Compensation, it doesn't matter; the point is you tell the camera to make the image darker on purpose, so that the camera has a chance to capture more of the highlights. So, let's try that:
Raw image, under-exposed by 1 stop (half as bright as the original). The sun is still over-exposed, but it is a lot better. The ground, though..
Now we have another issue; the ground is way too dark. Depending on your camera, it might not be a problem. Many modern cameras (Nikons especially) have amazing shadow recovery that let you get the detail back when processing a RAW file. But many cameras don't, and in this image, my trusty Nikon doesn't fare too well either. Let's try increasing the shadows and taking a detailed look at what it does.
Shadows at +100, nothing else. Too good to be true?
3:1 magnification of a shadow area. See all the noise? That pixelated mess is very typical when you raise deep shadows on most cameras.
So, we got the detail back, at least when looking at the image from a distance. But we paid heavily. When we increase the shadows, we are asking Lightroom to amplify the brightness in an area where our camera was struggling to get enough light. As a result, we see very clear noise in our image. Not good if we wan't to publish a high-resolution image online, or hang a print on our wall.
So, finally we arrive at what HDR actually is. To fix the problems we are having, we need more dynamic range. We need more information in the highlights and the shadows. Using HDR, we accomplish this by taking multiple images and combining them to a single, data-packed image. We take images that are darker so that we get a nice sky with all the detail preserved, and we take images that are too bright, to get more data for the shadows, reducing noise in those areas. Then we feed all of them into the Lightroom HDR tool and let it combine them.
You can take as many photos as you like, but typically you would use 3 or 5 shots. The middle one is 'correctly' exposed, or at least what your camera thinks is correct. This will be your average. The other ones are spaced out both ways, with a difference of for example 1 stop. So with three images would have values of -1, 0 and +1 stop, and five would have -2, -1, 0, +1, +2. I took three with a difference of 2 stops (-2, 0, +2) because of the big range I was dealing with; the more images you take the better the end result will be, but the longer it takes to process. It helps to use a tripod when taking the images, but it isn't strictly necessary. Just remember that your bright shots might be 2 stops over-exposed, lowering your shutter speed by a factor of 4. In the series below the bright shot is at 1/6th of a second, making it tricky to hand-hold. I just used a tripod so that I din't need to worry about it. I'm lazy that way ;)
-2 stops, 'correct', +2 stops. The left one has a nice sky, the right one has no noise in the shadows, and the middle one is a good average.
Select the three images in Lightroom (e.g. using the ctrl key), right click, and choose 'Photo Merge' > 'HDR'. Lightroom will build a preview and show you some options.
Let's go through the options real quick.
- Auto align: Always tick this, even when using a tripod. It makes sure the images are aligned, pixel-by-pixel.
- Auto Tone: Leave this off, we want to process our image ourselves. Leaving it on is OK though, as you can reset those changes later if you don't like them.
- Deghost amount: If your image contains moving subjects (people, vehicles, moving branches, etc.), chances are that those objects have moved between shots. This setting lets Lightroom detect that movement, and remove it. I typically use 'Low' or 'Medium', but with this particular image Lightroom was misbehaving and only let me choose 'None'.
When ready, select 'Merge'. Lightroom will start the merging process. Note that the original images will not be touched; instead, a new file will be created. Lightroom can even merge RAW files, that external applications typically can't, and produces a HDR RAW file which is very convenient. Here is the new file, opened in the Develop module:
Doesn't look like much, does it? There are changes, if you look closely; the sky is a little bit darker, and the ground is a little bit lighter. The beauty, though, is in the data. Even though Lightroom doesn't tell you, it has a lot more range to play with now. Let's start opening up the shadows and checking the noise:
The noise is completely gone. Good riddance.
You'll see the same effect when processing the highlights; the detail is now back. From this point, you can edit the image as normal. I dropped the total exposure, added contrast, clarity and vibrance, and cropped some of the top and bottom (they weren't that interesting, and now the image is 16:9, fitting nicely into e.g. a Full HD screen).
Note how I cropped the leaves in the corner; you don't want distractions in the corners. The horizon is also close to a 3rd-line.
The I added some more vibrance to the overall image (from 15 to 30), and created a graduated filter to darken the sky a tiny bit, also adding a hint of saturation. The last thing was removing all the dust spots I had on my sensor/lens; go to 'Spot Removal' and turn on 'Visualize spots' and play with the slider until you clearly see the spots. Then just click each of them.
3 spots down, one to go (top right of the image).
And there you have it. I hope this tutorial helps you when you stumble on a great scene, but your camera doesn't want to play. Using the HDR function in Lightroom is really easy, as long as you remember to take the additional pictures. With it, you don't need the latest and greatest camera to produce amazing, high-range photos.
Next up, creating Panoramas in Lightroom, see you then :)
