HDR stands for High Dynamic Range and you most likely have seen many examples of this type of photography. There are may fantastic examples used in every outlet of photography, from commercial advertising to fine art prints. The there are the not-so-fantastic examples where the colors seem to clash and the image looks horribly over processed.
But what is Dynamic Range and how do we recognize it when we survey a scene? Hopefully this article will give you a little insight into what HDR is and how both our eyes and our camera interpret it.
Dynamic range, in its simplest form, is the ratio between the largest and the smallest values in a given range. In photography we are talking about light therefore it is the range between the darkest shadow to the lightest light. Because our eyes have millions of light receptors it is like having a finely tuned million megapixel camera in our heads. When we view a scene we can clearly discern all the subtle shades and colors. We can also discern a wider range of colors than a camera sensor since our brain can handle that kind of processing power.
A camera sensor in comparison has significantly fewer light receptors and a drastically weaker processor. Therefore its interpretation of how it sees light is much more limited. We further compound that with our display and printing limitations and you have now significantly reduced the tonal range being displayed. A simple example is in the graph below where a range from dark to light has been compressed into seven stops. Again realize that this is an oversimplification. But not to get off track.
In the full range from dark to light, our eyes can see a dynamic range of something like 16+ stops of light while a camera sensor or film has an approximate dynamic range of seven to nine stops. A significant smaller range. So what happens to these stops outside that full range our eyes can see? The answer is actually in your histogram--they get clipped.
Let us assume that you meter a scene, adjusting your metering to capture the light areas of the scene (i.e. sky, clouds, light colored shirt, etc.). As you can see in the diagram at right, the range captured leaves a lot of the shadows in limbo. The green bracketed area is your camera's dynamic range for the scene you're looking to capture while the red is the area outside your sensor's capabilities. Since those shadows now lie outside the dynamic range of the sensor it converts, by default, all that information as dark tones. Pulling up the histogram for that image reveals a graph similar to this one. Just what you feared--a lot of clipping in the shadows.
The opposite is true when you meter for the shadowed areas. In this case the meter favors the dark tones leaving all the light tones hanging. Just like with the previous example, all the light tones that fall outside of the range get compressed into the highlight area of the spectrum. Again, checking your histogram will show a graph pushed into the highlight side of the scale.
There are a couple of things to remember here. First, digital sensors tend to favor shadows more than highlights. What this means for you is that you have a better chance retaining more information in the shadows than in the highlights. In other words, The closer each extreme is to the end of the dynamic range, you will loose more information from the light side than from the dark side. Hmm, sounds a little like a Star Wars reference here. Second, once lost it is difficult to impossible to reclaim image data. Be careful when metering to make sure you are not clipping too much.
Let's get back to the beginning here, HDR or High Dynamic Range. While there are several ways of obtaining or expanding on a camera's dynamic range, the most commonly practiced method is to use several exposures and then stack them or tone map them into a single image that utilizes the full range from dark to light, without clipping. Unfortunately we are still limited by our digital sensor in capturing all the colors and subtleties and by our output technology and its ability to recreate what is captured.
Hopefully you now have a better understanding of what dynamic range is and how your camera interprets data.