© 2010 - 2012 Photography Art Cafe. All Rights Reserved.
Introducing The Histogram - An Essential Tool For All Photographers
Over the last few lessons we've been learning how to use camera control to create well exposed pictures, with the visual characteristic you intend. Now let's take a look at a really useful tool to use when reviewing your images, to decide how you've done and whether you need to change the settings. It's the called the histogram and is extremely simple, but extremely useful.
What is the histogram?
The histogram is a simple graph that represents the tonal range of a photograph. It maps all the tones that have been captured in an image, displaying pure black at the far left and pure white at the far right.
The histogram can be easily accessed within the camera, and you can set things up in the menus so that it can be viewed whenever you playback your images. But why not just review the tonal range of your images by actually looking at them on the LCD!?
Well, the histogram is there to supplement the LCD's image playback, but it's valuable because the LCD screen is hardly the best way to view images. It's small and, if you use a plastic cover, can be quite scratched and dull. So combining the two is a good idea.
The histogram gives you a really useful snapshot of the exposure of an image in graph form. The higher the peaks in a given area, the greater the amount of tonal information of that kind in the photo.
For example, if the graph is bunched up in peaks over to the left the photo must be pretty dark, and if there is a dominance of imformation on the right it must be bright.
"Clipping" - overexposed and underexposed areas
One of the best uses of the histogram is as a guide to where you have overexposed or underexposed a picture. Sometimes this can happen to such an extent that in the brightest/darkest areas of an image all the information has been lost, and all that remains is pure white or pure black. This is called '"clipping" of the highlights/shadows.
When clipping has occurred, the histogram will display information squashed right up against the far left (for black) or far right (for white) of the graph in a straight vertical line. When you see this, it usually means you need to adjust the exposure to reign in the tones at the extreme ends. But, of course, sometimes clipping is unavoidable or may even be part of the high key/low key effect you are going for.
What should a histogram look like?
There is no such thing as a perfect histogram. The tonal range of a photograph is entirely up to your creative preferences. But, as a rule of thumb, when trying to create well exposed pictures in most situations it's good to see a fairly wide spread of tonal information, instead of all the tones bunched up in one area.
Not only does a wide range of tones suggest a well balance exposure, it also gives you extra flexibilty in post-processing to make adjustments. With more tonal information you can stretch things more before image quality starts to degrade.
Exposing 'to the right'
Have yo uever heard photographers talking about "exposing to the right"? This is a reference to the way a photo looks on the histogram. In other words, exposing to the right means capturing more bright tones than neutral and dark ones.
This histogram shows the majority of information is fairly bright (to the right), but there is also quite a nice range of tones.
There's quite a lot of benefit in exposing slightly to the right. Modern digital cameras are capable of recording more information in bright areas than dark ones. Plus, when it comes to post-processing, reducing brightness can easily be done without reducing image quality. Brightening dark areas, on the other hand, very often introduces noise (graininess).
Examples of histograms
Let's take a look at some actual examples of histograms and what they tell us about the pictures they represent:
The above histogram shows the majority of information over to the right of the graph (an example of exposing to the right). This tells us that the image is very bright. In fact, we can even see the highlights have been slightly clipped with the straight line on the far right.
In this case, it was necessary to slightly clip the highlights to make the rest of the image bright enough, and it doesn't harm the quality of the photo. Had the clipping been much bigger I would have changed the exposure triangle settings.
Unlike the first histogram, the information on this one is mainly on the left, which suggests an underexposed photo. There is almost no information on the right hand side of the graph.
This means that there will be littlle clear detail in the image and not much contrast either. It was definitely necessary to change the exposure settings for this image and produce a brighter photo.
This histogram is different to the other 2 in that it shows a very wide range of tonal information. The largest peaks are towards the far right and far left, which shows there is strong contrast in the image. This is the kind of shot that would work well in black and white.
Good picture/'bad' histogram:
Finally, this histogram is a bit unusual. In most situations it would look all wrong, with severe clipping of the highlights and shadows and a dominance of information in the dark shadows.
But the picture actually turned out pretty well as intended, which shows that how 'good' a histogram is really does depend on the subject. This image features a silhouette of an archway and a tree disappearing into bright mist, both of which needed to be quite extreme tones.
Histograms in Photoshop
Histograms come in handy when reviewing pictures during a shoot, as well as when working on them in post-production. As you make adjustments to images in Photoshop or Lightroom (or whatever) by keeping the histogram visible you can easily see whether your changes have caused any clipping.
Even more usefully, histograms in post-processing can be a guide to when your adjustments have lead to something called "posterization". Posterization occurs when the tones within an image are stretched so far (often by Levels or Curves adjustments) that they are forced to cover a broader range than was present in the original image.
This effect produces sharp spikes on the histogram, with slight gaps between them, instead of the normal solid graph of information. When posterization happens you see 'stripes' or 'bands' across the image, where there is not enough tonal information.
Histogram for an image with posterization:
Posterization in an image:
So if you see posterization in the histogram, it's worth going back and re-editing your original image with less drastic adjustments, that don't try to stretch the tonal range too far.
So the histogram is a great tool for assessing the exposure and tonal range of your pictures. Used in conjunction with the LCD preview, it gives you an extra insight into photos during a shoot. On the whole it's good to capture as much information as you can in a picture, without clipping and, sometimes, with slight bias to the right. But there are no strict rules!
In post-production, keeping the histogram visible lets you easily see whether your adjustments have harmed image quality, with clipping or posterization.