I'm pleased to say that in this post I won't be discussing depth of field equations or pupil magnification. This post is all about how I post process my deep focus, high dynamic range landscape scenes. That is a scene that forces me to focus and exposure bracket.
For the focus element, it doesn't matter too much how you arrived at the points of focus through the scene, eg: some cameras offer an auto focus bracketing mode; others, like me, can use in camera scripting via Magic Lantern or CHDK; finally some will manually focus bracket, even guessing the points of focus.
For the exposure element, many will exposure bracket at each focus station, thus generating a lot of brackets to process in post, eg 5 focus stations x 3 exposure brackets = 15 images to process.
Once captured, many will then 'throw' the images at specialist software to first create an 'HDR image' at each focus station, then take these processed HDR images into a focus blending piece of software, eg Helicon Focus.
Personally, I have evolved an alternative approach for deep focus and high dynamic range landscape photography, that I call sky bracketing.
In this approach I take one exposure image for the sky at 'infinity', and then a set of focus brackets for the land, at a different exposure to that of the sky.
For the sky exposure I will ETTR (Expose To The Right) using the histogram and not worry about the land being underexposed.
Having captured the sky, I will then adjust the exposure, for the land focus brackets, by the required Ev, ie so the shadows in the land are not blocked. You could use an in-camera spot reading or use an external spot meter; but the live view histogram is the best tool.
If you are lucky enough to have a Canon running one of my scripts, all he above is handled automatically, once you have ETTRed for the sky.
However you generate your bracket set, in the end you will arrive at something looking like this:
Here we see four focus brackets for the land, and one exposure bracket for the sky at infinity. All, of course, taken at a fixed WB, ie never use auto WB when bracketing ;-)
In the above example I ETTRed for the sky and used a 3Ev uplift for the land, thus ensuring I captured the tonal details in the shadows. The two histograms looking like this - top for the sky, and bottom for the land:
Clearly showing that, from an exposure perspective, we have covered ourselves.
From a focus perspective we see four images, ie one at about 3 * hyperfocal (H) = an infinity blur of about the CoC/3, one at H, one at around H/3 and one at around H/5.
The first step is to adjust the exposure of the focus brackets in Lightroom, ready for focus blending in Photoshop, noting the sky is blown out, but the foreground exposure is now perfect:
These four images are then sent to Photoshop as layers in a single image; and aligned and focus blended in Photoshop; giving us a final, focus stacked image, for the foreground - accepting the sky is blow out in places:
Note that in this example I used the Photoshop auto blending feature. However, as focus will be in bands through the scene, I could also have tried focus blending with manual masks.
The next step is to return to Lightroom and adjust the sky bracket's exposure, so it better matches with the foreground, focus stacked image. Once this is done, then select 'Edit in Photoshop', which will result in you have two images open in Photoshop: the focus blended foreground and the sky exposure, taken at infinity.
All we need to do now is flatten the focus stack, and bring the sky bracket into a layer on top of the flattened focus stack. Add a mask to the sky bracket and an exposure adjustment layer, eg using curves or levels, to the focus stack, to help blend the exposures of the two image layers:
Although one could use a luminosity mask, the blending of the two images usually only requires a simple mask as shown below, eg:
Having completed the Photoshop part of the process, I then flatten the image, carry out any final Photoshop editing, eg cloning etc, and then return to Lightroom for final tweaking.
In this case the final deep focus, high dynamic range image looks like this:
Bottom line: I hope I've showed in this post that high dynamic range, deep focus photography is a simple process, both in the field and in post processing. It also doesn't need any special software, other than one that can handle blending (auto or with manual masks).
As usual I welcome any feedback on this post.