fbpx

Histograms Demystified!

Are you using your histograms? Do you even know what this is? The histogram is one of the greatest advancements in photography and one of your must useful tools on your camera. Despite this fact, I have found that many beginning photographers tend to shy away from their histogram like the plague! If this is you, or if you are still not incorporating the histogram into how you shoot, than you are holding yourself back. Its the 21st century and you need to photograph like it!

Basically your histogram is the modern day light meter. This little graph will tell you how many stops of light your camera is capable of capturing detail, where your exposure falls across this range, if you are over exposing, underexposing, etc. . . Your histogram will even break down your exposure into the three color channels of red, green, and blue – which is far more helpful than the single histogram that most photographers check if they even check their histogram at all. So your histogram is all about exposure and is the tool that will tell you if you nailed it or not. Getting the exposure right in the camera (not by adjusting a slider in RAW) is crucial to producing the highest quality image that you can.

When it comes to exposure, you have to understand that your camera can only record information across a limited range of light. In photography we refer to this typically as stops of light. Though your eyeball may be able to record 12 on up to potentially 20 “stops” of light (think tones of light from black to stark white) your camera can only record 5, and thus you need to keep that information within the tonal range that your camera can record! This is why what you see is never going to be what you get exactly. Your eyeball, and the brain that is full of thoughts, emotions, memories, etc. . . that interprets what you see is just a couple million years ahead in technological advancements from that of your digital camera (regardless of how much you paid for it!). Being able to keep your exposure inside of this range of light is the first step to nailing exposure.

Histograms are set up from from absolute black to absolute white (left to right). This means that if your histogram is squeezed up against the left hand side of the graph, than you have underexposed your image and there will be areas of complete black containing zero information. Likewise, if your histogram is cut off at the right side of the graph, than you overexposed and will have parts of your photograph that are blown out white.

This does not mean that you should have a bell curve in your histogram. There is no proper shape to what your histogram is supposed to look like. I repeat, there is no proper shape to a histogram!

Every scene is different. Every scene will offer a differing amount of lights and darks, shadows and highlights. So this means that your histogram will look different for every scene. You just need to remember that the left side is dark, the right side is bright. “Sayings” often help us remember stuff… so think right is bright. Or, right is white. What ever works for you here.

Now, in terms of the different histograms that your camera will display, you should have 4. The most important ones for you to consider here are the RGB histograms (red, green, blue). Why? Just because the single histogram shows that you have kept your exposure within range, this is only an average of the RGB. Looking at the individual red, green, and blue channels themselves will often reveal that you have blown out your reds for instance while the single histogram shows that everything is within acceptable range. Why does this matter? Say this image is a sunrise and the dominant colors of the sky are red. Now, instead of depth to the color of the sky, you will end up with a section that is noticeably flat and potentially blotchy.

Now if you are still with me thus far, I might be getting ready to enter into territory that may get confusing. There is a saying (I like sayings) when it comes to histograms. . . “expose to the right.” If you remember this much than you will be doing good here. Explanation however, is in order though.

You must first understand that each full change in f/stop gives up a doubling effect. What the heck does that mean? Well, it means that if I open up my f/stop from say f/11 to f/8 than I am effectively doubling the amount of light that enters in through the lens and to the camera’s sensor. That part is easy enough. But because of this doubling effect and they way that the camera’s sensor is designed to record light, the brightest 1/5 has the best ability to record the greatest amount of information.

Here is where it gets weird.

RAW files are going to capture about 4,096 distinct tonal values. Since f/stops and your camera’s sensor works on a doubling effect of light, this means that the brightest 1/5 of the histogram will contain 2048 tones, the next 5th 1024, and on down to 128 tones on the dark side of the scale. Thus, the vast majority – a full 50% – of the cameras ability to record information occurs within the brightest 1/5 of your histogram. Check out Adobe’s site for more information on this ( http://wwwimages.adobe.com/www.adobe.com/content/dam/Adobe/en/products/photoshop/pdfs/linear_gamma.pdf )

Moving away from math and algorithms what this means then is that if your histogram is bunching up near the left hand side of graph, than you will be contending with more digital noise, and less to work with in the image. 

This is a rule of thumb of course. And as with all creative rules, they are most certainly meant to be broken! Its just that we have to understand why the rules are there to begin with and how that they work before we can successfully break them!

OK, so key highlights of histograms….

1. use them!

2. left is dark, right is bright

4. There is no such thing as a proper shape to your histogram

3. when the information is cut off at the edge of the graph (left or right) this means that you have lost information and will have absolute black or absolute white instead.

4. In most instances, expose to the right of the histogram without overexposing to avoid noise and record the maximum amount of tonal values.

5. Experiment! Get out and shoot. Try different exposure settings and look at how your histogram changes accordingly.

This entry was posted in Technical Skills.