One of the features offered by many cameras these days is the ability to display histograms. These can be viewed for a recorded image, or in many cases for a live scene, allowing you to judge the exposure of a picture. So what exactly are histograms?
A histogram shows a graphic representation of the balance of tones across a range from black to white, through the various shades of grey. It shows the relative number of pixels across the range, usually for the full spectrum of light, but often for just red, green or blue – the colours that make up a projected image.
Using a histogram you can judge the exposure of an image – whether it is too bright, too dark, or just right.
If an image is underexposed, most of the pixels will be in the darker range and there will be an absence of brighter pixels. This can be seen in the image below and the accompanying histogram.
This histogram shows us that most of the pixels are a darker grey. There are no pure white pixels (nor any pure black either). This shows a poor tonal range.
The photo below shows the opposite problem – an overexposed image.
Here we can see that most of the pixels are distributed towards the right of the graph. They are mostly a lighter grey or pure white. There are very few darker pixels. The fact that the graph is clipped to the right shows that the image is “blown out”. This means that detail has been lost at the top end of the tonal range.
An ideal histogram would have a reasonable balance of tones, through from pure black to pure white, with a peaks and troughs across the greys between. Having said that, there is no such thing as a perfect histogram. Certain types of photograph will give a different profile. High contrast scenes for instance will feature peaks on the far left and right of the histogram. Other exceptional lighting conditions will also produce unusual but perfectly acceptable histograms
Whilst the above explanation is much simplified, it does give a good introduction to histograms.