HDR Photography Workflow / SML Tutorials
This presentation briefly goes through the steps of how I processed a high dynamic range image of a cloudscape during stormy season in Hong Kong.
High dynamic range images or HDR, is the process of combining multiple low dynamic range captures into a single image, when a single capture does not afford the necessary range for what is intended.
Three captures were made for the making of the final image. All three images share the same ISO, the same aperture value, and the same focal length. They differ in shutter speed, and the result is shown here, with one normal exposure (0EV), one 3 stops down (-3EV) and one 3 stops up (+3EV).
The image on the left was exposed for the overall intent: the clouds and the reflection on the sea. The image in center was exposed for the highlights, which provides some tonal details in the sunlight in this case. And the image on the right is exposed for the shadow areas, where the buildings on the horizon are now correctly exposed, and you can see the patterns on the water surface.
After importing my RAW captures into Lightroom, I have exported them to merge to HDR via Photoshop. I used to use Photomatix for this but lately I have been using Photoshop because it was easier for my workflow.
The important bits to note here is that you should be focusing on extracting details from the captures, and not trying to create the final image. By keeping everything in 16-bit color space we can manipulate the image further once we have all our materials together.
Most strange looking HDR images found on the web is a result of people exporting to JPEGs after this. But this image was not what I had in mind when I looked at the landscape, so let’s work on this further.
To further reveal details in the shadow, I created a curve which targets only the shadow area. Shown here in the layer blending options you can see that I have targeted the curve adjustment to only affect a graduated area. If I do not do this, the sunlight would have been over-exposed again.
Also note that I am using the Lab color mode inside Photoshop. This allows me to work with the luminance, or lightness, separated from the color a/b channels.
After I am satisfied with the shadow overall contrast, I work with the overall luminance contrast for the overall image. Obviously the curve shown here is specific to this image.
Now I created a layer mask with a basic gradient fill so I can use the curve to adjust the tonal contrast for the sky.
Then I do the same for the water area. Note that at this point this image is very similar to by 0EV capture—no surprises there. This was what I saw when I photographed this photo. I do HDR not for effect, but just so that I can have some details on the shadow area.
At this point you see that really there is not much color to this scene. In fact, they stand to distract the overall image, so I removed it. Here since I have been working in Lab mode I just quickly removed the color channels or just removed the color saturation. If I work with images filled with color I usually would use the channel mixer in RGB mode or use the black and white tweaking modules inside Lightroom. But as you can see there is not much color in this image to start with so I can simply just remove the color saturation.
After importing back into Lightroom, I tweak things further as I feel that I can still get some more details in the shadow area. If I were inside Photoshop I would do a image-wide shadows/highlights tweak but since Lightroom version 4 there is this new Clarity parameter which works very well in manipulating local contrast so I used it.
And that’s it. But remember, every single image is different. This is not a recipes. There are no rules. Now go have some fun!