Did you know the human eye roughly has a 24-stops dynamic ? Well, that is about 2 times more than the best professional camera.
This explains why you are able to see more of a place where darkness and light combine with the naked eye than when you attempt to photograph it. A camera will require multiple exposures at different exposition parameters to render a highly contrasted scene. This is what HDR (High Dynamic Range) photography is all about.
Many software components, internal to your device, or part of a post-processing tool, allow you to automatically combine multiple exposures into one HDR picture.
I personally prefer to control the whole process, judging by myself what zones need enhancement, adding in my own sensitivity to the result. Besides, some pictures are very easy to combine with a simple layer technique.
Tips to get the shots
– Use a tripod, or at least stabilize your camera so that the shots will be stackable.
– If available, use the multi-exposure function of your camera. Otherwise select the semi-automatic aperture priority mode and manage the exposure scale manually.
– Choose the average metering mode (“[ ]”), so that the metering will not favour any zone in the frame.
– Take as many shots as necessary to cover all the exposure range (a multi-exposure function may limit to 3 or 5). No over- or under-exposed zone must be left. Each shot should be separated by a same exposure value (EV), ideally comprised between 1 and 2EV
In this example I got 3 separate photos, taken at -2, 0 and +2EV. This means the first shot was under-exposed by 2EV, the second one correctly exposed, and the third one over-exposed by 2EV.
You can see the multiple-exposure was necessary, since the camera could not expose the inside of the cave, too dark, and the outside, too bright, within a same shot.
Start by opening all the photos as layers within the same project in Photoshop. Stack them from the more exposed to the less exposed.
I am going to work with layer masks. This will allow us to discard specific parts of the layers. In this example, I wish to discard incorrectly-exposed parts.
Select the less exposed layer (the first one), and create a mask.
One of the primary properties of a layer mask is that it links the layer to a transparency scale based on shades of grey : any “white” part will show the corresponding zone with 0% transparency, any “black” part will make it 100% transparent, letting the layer from under show.
A selected mask can be painted with any painting tool (brush, gradient, paint bucket…) but only accepts greyscale (white – grey – black) colors.
I will use these mask shades to select what zones of each layer to keep.
This first base image has the right exposure towards the outside of the cave but the inside is too dark, so I’ll use a black brush to exclude any under-exposed zone. The brush hardness must be low, so that the excluded zones won’t clash with the non-excluded ones.
If I disable the other 2 layers’ visibility, you can see what is left of my first layer :
The cave entrance and the area surrounding it are now just fine, but the inside of the cave is still too dark, so I will mask the second image to reveal parts of the third layer.
And voilà !
Some adjustments regarding levels and colors may be necessary, but the base composition is now complete.
This technique is just fine for compositions where the different exposure zones are easily distinguishable.
If the different zones were too intertwined, the masks would be a lot more complex to detail subtly.