Learning never exhausts the mind
Home >  Photography > Tutorials > Understanding Histograms

Published 1st December 2017 by

It's just a simple graph, but the histogram has had a major impact on how we go about exposing our photographs on location and processing them later.

What is a Histogram?

A histogram is a graphical representation of the tonal values of your image. In other words, it shows the number of tones of particular brightness found in your photograph ranging from black (0% brightness) to white (100% brightness). Histograms have a vertical axis showing the number of pixels in each of the 256 brightness value channels, and the horizontal axis represents the brightness of the light. The latter ranges from pure black on the left to pure white on the right. This range of brightness is often split into the broader categories of shadows, midtones and highlights. How the histogram looks relies entirely on the scene it's representing, so there is no one, ideal histogram shape to aim for.

ERROR: Image id 7500 is invaid.

When exposing for landscapes, there are two rules you really need to think about. The first is to expose for a midtone and the second is to avoid burnout.

A midtone is an area of the scene that falls halfway between the darkest and lightest parts of the shot. Once located, you can use the midtones as a reference point to ensure that the midtones in the scene translate to midtones in the captured image. It's a simple concept but ensures that the foreground details are clearly visible for a natural looking picture. Grass and other foliage make a reliable midtone, so long as it's under the same lighting conditions as the majority of the shot.

Burnout occurs when highlight tones become so bright they turn pure white - all underlying detail is irrevocably lost. A spot of burnout around the sun is perfectly OK, but huge stretches of a boring white sky are not. Ideally, you want all the major parts of the scene to contain plenty of colour and detail.

Avoiding burnout in the sky, while getting a good foreground exposure, is easier said than done. The reason being that the vast tonal range of the landscape is often too much for the camera to handle. it can only capture a slice of what's on offer, so something has to give. If you expose for the midtones then there is a good chance the sky will burn out. If you expose for the highlights, then the foreground will fall into silhouette. You need to think carefully about what's most important in your scene and expose accordingly. of course, if you don't want to compromise you can use a neutral density or graduated filter.

Histogram Clipping

ERROR: Image id 7352 is invaid.

Highlights or shadow clipping can occur when the brightness values in a scene, based on the exposure settings used, fall outside the histogram range. There is nothing you can do post-process to recover the detail in those areas of the image. Shooting in RAW does offer some latitude for recovery, but the clipping has to be slight for there to still be detail. For this reason, it is important to view the histogram in the camera on the image preview screen and correct the shot in the field, rather than attempt to fix it later.

To avoid clipping, ensure that your histogram barely contacts the left or right sides of the graph. That will ensure that you retain the details you need in those dark and light areas.

Some camera feature a highlight warning which is often called "blinkies". This can give you a visual warning when playing back your shots when clipping occurs. The exact method varies by manufacturer, but in general, highlights will appear to blink between white and black. If you see these blinks then clipping has occurred and you need to address that.

How to Read a Histogram

You can call up the histogram on your camera's LCD to judge the tonal distribution in a photo. The horizontal axis show pixel brightness, ranging from pure black on the left to pure white on the right. The vertical axis shows the number of pixels at a particular brightness level.

ERROR: Image id 7359 is invaid.

ERROR: Image id 7349 is invaid.

ERROR: Image id 7350 is invaid.

Under Exposed

Not enough light has reached the sensor, either too fast a shutter speed or too narrow an aperture, or both. This shot's histogram is pushed all the way to the left, a situation known as clipping. If printed, the clipped areas will appear completely black and all shadow detail will be lost.

ERROR: Image id 7353 is invaid.

ERROR: Image id 7354 is invaid.

Correctly Exposed

This histogram should correspond with the tones of your scene. A dark scene should have a histogram with a bell shape on the left. A light scene should have a histogram with a bell shape on the right. For daylight scenes, expose the scene as far to the right as possible without clipping the highlights.

ERROR: Image id 7347 is invaid.

ERROR: Image id 7348 is invaid.

Over Exposed

Too much light has reached the sensor. Notice how the histogram reveals that the tonal distribution is pushed all the way to the right and that the highlights are clipped. If printed, the areas of the scene which should reveal detail will appear as pure white.

High Key, Low Key and Contrast

ERROR: Image id 7351 is invaid.

ERROR: Image id 7356 is invaid.

High Contrast scenes create a histogram that spikes at both the shadow and the highlights ends of the scale with fewer values in the midtones area.

ERROR: Image id 7355 is invaid.

ERROR: Image id 7360 is invaid.

Low Contrast scenes show the majority of brightness in the scene sits in the midtones region, with very few pixels at the highlight or shadow ends of the histogram.

ERROR: Image id 7358 is invaid.

ERROR: Image id 7357 is invaid.

If the majority of brightness values sit at the highlight end of the spectrum, the images is often referred to as being High Key.

ERROR: Image id 7346 is invaid.

ERROR: Image id 7345 is invaid.

When most of the values at the shadow end of the histogram is sometimes referred to as being Low Key.

The "Perfect Histogram" Myth

There are claims on the internet that one or another histogram shape should be perfect and that it is the shape to aim for in all your photography, the perfect histogram.

I don't believe this to be the case. There are many variables can affect the photo's histogram. The camera settings and the scene you're shooting directly affect your histogram before you even click the shutter. You can see from the examples above, different scenes and lighting create vastly differing histograms. Are they all incorrect? As long as there is no clipping and the midtones are sufficiently exposed then the perfect histogram is the one in which the photo looks good. Don't try and fit your scene, lighting and settings to make the perfect histogram shape. Let the scene speak to you, and if it requires more shadows, let the histogram reflect that; if it requires more highlights, that's fine too!

Leave a Reply

Fields marked with * are mandatory.

We respect your privacy, and will not make your email public. Hashed email address may be checked against Gravatar service to retrieve avatars. This site uses Akismet to reduce spam. Learn how your comment data is processed.