CameraPorn

Howto: Create Higher Dynamic Range With Bracketed Exposures

For those of you that are familiar with our recent blog project “Revisit and Retouch” you’ll recognize this image. I provided CameraPorn readers with a set of three bracketed exposures to have a go at creating their own unique version by any means necessary. The image above is not only my entry to the project, but also an exercise in retouching that focuses on the common practice of compositing multiple, bracketed exposures to create a final image that better represents the scene as viewed by the human eye… In more ‘technical’ terms, an image with a higher dynamic range.

Before getting into the nitty-gritty of how I came to this final image, it’s important to understand what I mean by “higher dynamic range.” Many of you are probably thinking, oh yeah, he means ‘HDR’ imagery like what you create with a program such as Photomatix and see plastered all over flickr, but you’re wrong. As neato as those images can look, they are rarely executed in a way that brings the final image to appear as the scene actually looked to the human eye, which in my opinion, is where the true value lies in creating higher dynamic range images. Creating a tone-mapped HDR image in an HDR program usually leads to oversaturated, dream-like and haloed images which look pretty cool but can be created by almost anybody with Photomatix and some bracketed images with little to no skill involved in most cases. Now don’t get me wrong there are some incredibly talented HDR ‘artists’ out there whose work amazes me, but the general image you see on flickr looks too fake for my tastes, and besides today we’re talking about using good old Photoshop and Lightroom. To see how I did it and learn a bit more about dynamic range, read on…

What is Dynamic Range?

Put simply and in relation to photography, its the range of values between light and dark areas of a scene. To make this a bit more clear, its first important to understand that the human eye can pick up a much wider dynamic range than a digital camera. If you’ve ever been shooting a sunset into the sun while it is still above the horizon for example, and to your eye the scene looks perfect so you take the shot, more than likely one of two things will happen. Either the clouds will look amazing while the foreground looks dark, or the foreground will be exposed correctly and the sky will be completely overexposed. To put this into context take a look at these two unretouched shots taken in the Florida Keys.

Notice that the shot on the left, though overexposed as a whole, retains detail in the trees and lighthouse, yet the cloud detail in the sky is completely blown out, then the image on the right shows great cloud detail, but the tree and lighthouse detail is lost.

To my eye at the time, it looked like a gorgeous shot, but the camera did not pickup the scene as I saw it because it cannot capture the same dynamic range my eye sees in a single capture.

How Can I Create Higher Dynamic Range?

Aside from having absolute perfect light, which we all know is rare, to produce a shot that appears even close to the dynamic range as seen by your eye, you need some help. A shot like this should contain detail not only in midtones, but also in highlights and shadows, and in most cases, especially high-contrast scenes, it won’t happen without the right tools.

Working With Graduated ND Filters

The traditional method for keeping shadow detail without totally blowing out the sky is to use filters, and more specifically, graduated neutral density filters. Back in the film days before Photoshop and Lightroom, if you didn’t get the shot on the film, salvaging it in post-processing was not the possibility it is today, so graduated ND filters were a necessity for landscape photographers. Think about them as partial sunglasses for your camera, and check out the picture on the left (photo credit: Dave77459 via flickr {cc}) and you should get a pretty good idea. I’m not going to go into further detail about using Graduated ND’s as using them is not the point of this article, but I won’t leave you hanging. Check out these articles on Digital Photography School and Wikipedia for a more detailed explanation.
Getting back on topic, in the digital age of today, these filters are not the necessity they once were, because we can now use bracketed exposures and a little post processing magic to gain similar and even better results.

Working With Bracketed Images

First, for those of you who don’t know, I should explain what a bracketed image is. Most of today’s DSLRs have a function called Auto Exposure Bracketing (AEB). AEB in conjunction with high-speed shooting mode, takes three, five, or more images in quick succession at different exposure intervals. AEB is useful in tricky exposure situations where speed is key, to take multiple exposures of the same frame and be sure you got the right shot. In this case, we are using it to capture the same high-contrast scene in order to expose for highlights, shadows and midtones with the plan of compositing these shots together later to create an image with a higher dynamic range that better represents what your eyes saw when the shot was taken. Many of you are probably familiar with this concept after reading one of the many HDR tutorials floating around the web, and its the same method you would use when capturing an image you plan to drop into Photomatix, but in this case we are using Lightroom, Photoshop, and our retouching ability to build the best final image we can.

Revisit and Retouch: Grand Cayman Fuel Depot – How I Did It

The challenge posed to CameraPorn readers was to take the three bracketed exposures above and create the best final image they could. For me, the goal was to create something that not only represented the scene as I saw it, but to utilize the retouching skills I’ve learned in the year and half or so since I took the shot. Here are the steps I took:

1. RAW Processing in Adobe Photoshop Lightroom

I began with basic adjustments within Lightroom, my workflow tool of choice. The adjustments made here could also be completed withing Adobe Camera RAW (ACR). The adjustments consisted of:

A. Straightening the horizon, cropping and matching the white balance of each image.
This was accomplished quickly by adjusting just one of the images and using the ‘copy settings’ tool to apply the same changes to the 2 other images.

B. Adjusting Exposure and Contrast of Each image to reflect what I want out of each.
This step was a bit more time consuming. I first looked at the 0EV image, and adjusted it so the histogram showed as little clipping as possible. This way, when I brought it into Photoshop, I would have as much pixel data to work with as possible. I did this by making slight exposure and recovery adjustments. Notice that the histogram doesn’t touch the left or right side, meaning no black or white clipping.As you can see, the image isn’t looking that great yet, but thats ok. Read on….

For the -2EV and +2EV images, I wanted to adjust each only for the part of the image I expected to use when I composited the images together in Photoshop. For the -2EV image, this meant the sky and tanks. I adjusted exposure, blacks, and contrast until they looked as i wanted, ignoring how dark the water was looking because I knew the water would be coming from the other images after compositing. For the +2EV image, I adjusted so the water and rock shadows until they looked as I would want them to after compositing. I used the same methods, along with a bit of highlights and shadows to make the churning water pop a bit more.

C. Export To Photoshop CS3

I then used Lightroom’s integrated Photoshop editing launcher to send each of the images to Photoshop as 16-Bit Tiff files.

2. Stack in One File, Align Layers

So now I’m in Photoshop, with each of the three files open and I need to get them in one unified file. I drag the +2EV and -2EV images into the 0EV image and close the other two files. Next, I need to align the layers. The shots were not taken using a tripod so there is a slight variation in each image, but Photoshops’s CS3’s Align Layers tool makes it an nearly instant process.

3. Composite The Image

Here’s where the real work comes in. To begin, create layer masks for the +2EV and -2EV images.

Next, begin ‘painting’ with the layer mask active with a soft brush on each of the the masked layers to make them show what you want to see from each. For the -2EV layer, I paint over the sky and tanks with a white brush to make them visible. For the +2EV layer, I paint over parts of the water and rocks. Continue to do this until your image begins to look cohesive, and the different layers match. If necessarry, feel free to do some Levels or Curves adjustments to give everything a unified look.

4. Dodging and Burning

Before Dodging & Burning

Instead of using the typical built-in dodge and burn tools of Photoshop CS3, I used a non-destructive method that consists of creating a blank layer on the top of your layer stack. Set the blending mode to ’soft light.’ From there, you can use white and black soft brushes to do some non-destructive dodging and burning. I used this method to bring detail back to dark rocks, brighten the swirling water and tweak the cloud contrast slightly. My dodge and burn layer looked as follows:

And the result after dodging and burning:

5. Spot Removal

Yuck! First of all, to all of the participants in the project, I’m sorry!!!! I know this step must have been a big pain. There were big waves crashing up the seawall next to where I was shooting, and tons of sea spray got on the lens, but I have some good news for next time, there’s an easier way than atacking each spot with the healing brush! Next time, try using the Elliptical Selection tool to select each spot, feather your selection and use the Noise>Dust & Scratches Filter. All the spots, GONE!

6. LAB Color

This is a great method that I picked up from a variety of sources including Scott Kelby’s 7-Point System book as well as Brian Auer’s Epic Edits. For full details, click on Brian’s link, but basically goes like this:

• Flatten Image
• Convert your image to LAB (Image>Mode>LAB)
• Apply Image (Image>Apply Image)
• Play with the settings. I went with Soft Light Blending, Channel B, 70% Opacity.
  
• Once you get the result you want, convert your image back to RGB, and its time for some sharpening… the last step!

7. Sharpening

And we’ve hit the home stretch. You have your flattened image, now its time to sharpen. I used the Unsharp Mask, with the settings: Amount-70%, Radius-1px, Threshold-4 levels.But wait, not done yet. ANother nugget I picked up from the 7 Poitn System, sharpening can cause color halos and artifacts in color images, so directly after the sharpen, click Edit>Fade Unsharp Mask. Select the Luminosity option. This means the sharpening only occurs on the the Luminosity channel, leaving the color channels untouched.

And thats gonna do it.. we are finished and have completed an image with the appearance of higher dynamic range, without the use of HDR softwware cheating ! I hope this brief tutorial has been enough to help you bring more dynamic range and life into your images. If you have any questions, please leave them in the comments. For all you project participants! Expect to see the final results posted soon, and have your work voted on by the readers. Thanks again for participating and here again is the final image.