Digital Dynamic Range
Read MoreCourthouse Towers at Sunrise - Arches National Park, Utah Link To Original Image: http://www.tom-hill.biz/Galleries/Scenics/Utah/20957775_CZGg5B#!i=1726113116&k=stmCTvZ __________ If you're new to digital photography you might already know there are a million compromises to deal with when out in the field. It all has to do with the limitations of your equipment meaning your camera does not see what you eye sees, which might be a surprise to you. Today's topic is about one of those limitations. We're going to talk about dynamic range. Dynamic range is the breadth of luminance from very dark to white bright. This breadth is usually measured with a thing called Exposure Value (EV) or if you're old school, in F-Stops. Essentially the increase in brightness--i.e. luminance--on a dark gray object to one that's "twice as bright" increases it's EV by one or a single F-Stop. For two EV's it's a double double or 4 times the light increase. According to Wikipedia the human eye with it's brain involved and various other chemical reactions can discern about 20 stops or about a million times the light increase. I've heard the number about 11 stops, which seems to make more sense. Still it's a huge number. If you're wondering, your camera probably does about 5 stops meaning whatever your camera sees is not nearly what your eye can do. So begins the quest to get more than your equipment will allow. The technique we're talking about today is HDR. I have an article about what HDR means: (link: http://www.tom-hill.biz/Articles/Image-Processing/What-Is-HDR/21543728_ddX5sf#!i=1717604291&k=8NTVZcF) The basic premise of HDR is collecting two or more exposure values of the same scene then merging them together using software back at home. In the example I'm using today, I collected two sets of exposures separated by two stops--or four times the light. The technique I'm teaching here uses Photoshop and it's layers and mask features. If you're familiar with these you can easily understand what I'm about to introduce. The idea is you stack one exposure on top of the other then use a mask to mask out portions of the upper layer to reveal the lower one. The technique I'm teaching here is you'll be stacking layers with the bottom most the darkest layer with the upper one a bit lighter. In this case, the dark layer has the best highlights. The upper layer has the best shadows. The goal is to build a mask that allows the best highlights--the sunrise--to come through while preserving the best shadows.
Fig. 1 - Lowest Layer - The Brightest Layer
Here the bottom layer is pasted into Photoshop revealing the brightest portion of the scene. In this case the brightest layer provides the detailed information for the deepest shadows. The highlights (where the sun is rising) are blown out and lack detail.Fig. 2 - Second Layer - The Darkest Layer
There the next higher layer is the darker layer showing the best detail in the highlights (where the sun is rising). Unfortunately, the darkest layer has little in terms of shadow detail. The next step is to combine the two layers so the detail of the shadows and highlights are retained.
This is where a Layer Mask comes into use. Remember a Layer Mask is simply a mask that blocks out the portion of the layer it's masking to reveal what's lower. In our example, we'll be trying to block out the Darkest Layer's dark parts while retaining its bright parts. How do we do that?Fig. 3 - The Layer Mask
Here's how we do that. We first select the bright layer--the lowest one--and select the entire layer and copy it. Next, we make a layer mask for the dark layer and then option click that layer mask to turn it into black and white mode.
When in black and white mode when you paste into that layer mask whatever you copied is turned into black and white. The colors are unimportant when working with a layer mask. Remember a layer mask is simply something that blocks out the layer when where the mask is dark and reveals the layer where the mask in light.
In our situation, we're using the lower layer's tonal quality to create a mask for the next higher layer. Where the lower--i.e. bright layer--is most light--i.e. the sun--when pasted into the dark layer layer mask, the very detailed highlight area of the sun will be revealed. We'll see that in the next figure.Fig. 5 - Adjustment Layer Applied to Each Layer as a Clipping Mask
The technique is really simple. You can adjust the tonal qualities of each layer by applying a curve. So you can tweak things down the road I suggest making making an Adjustment Layer a Clipping Mask to that layer. Then, you can make little tweaks on your image without severely affecting the overall digital health of your image.
In this case, you can see the Adjustment Layer applied to the brightest layer--the lowest one--with a curve that generally makes the layer darker.
What you don't see in this example is the Adjustment Layer applied to the dark layer simply increases the contrast of the layer.
Remember, when using this technique of applying Adjustment Layers as a Clipping Mask, you're only affecting the layer you're making the adjustment to. You're not affecting both layers in this case. That will happen in the next pane.Fig. 6 - Overall Adjustment Layers
Now that you've tweaked the individual layers using Adjustment Layers as Clipping Masks, you're really adjusted the tonal quality relationship between the two layers. In the end if you've got everything just right, you might consider applying an overall adjustment that affects the entire image. It helps the overall effect of your image. In this case, I applied a curve that increased the overall contrast of the image and I applied a little vibrance which caused the saturation of some of the blues to be a bit more "snappy".
The end product is what you see at the top of this page. The adjustments to each layer are subtle yet important. Each layer gives a piece of itself to the overall image. The end result is a final image with substantially more apparent dynamic range. This effect tends to over-come the limitations of your equipment allowing what you print match more closely with your images.