Learning the Histogram

Learning the Histogram

To kick this blog off in spectacular fashion, let’s find out what a histogram is. The Histogram is a graph that tells you the brightness (or luminosity, we’ll touch on this later) of the pixels that make up your image. From left to right it measures black pixels to white pixels with shadows, highlights and midtones being in the middle. The furthest left point is pure black, while the furthest right point is pure white. The grey and coloured peaks between these points represent how many of each different pixel there is at those specific brightness or luminosity levels throughout the image. Simple, right? At a quick glance you can tell if you’ve captured all the pixel information in the scene in front of you.

A Histogram. It's just as afraid of you as you are of it.

Let’s break down the Histogram a smidge more! Understanding the histogram will instantly improve your photography believe it or not.

These are a general guideline for good exposure, however in photography there is always an exception to the rule!

The important thing to know about a Histogram is how the scale is measured. Notice I said it measures from black pixels to white pixels, not dark to bright. This is important because that’s how cameras see the world. They don’t look at bright sky and think it’s bright, they look at it and think ‘oh that sky is nearly white’. Obviously cameras can’t talk though, that would be ridiculous. The same can be said for a dark scene. To us we would say the scene is dark, but the camera will think ‘oh this scene is very black’. In order to judge how to expose your pictures correctly the camera will try place the black and white pixels on their correct place in the histogram. Often a camera will try and judge this by pushing dark pixels towards grey, and the white pixels towards grey.

Why? What do you get when you brighten black? Grey! What do you get when you darken white? Grey! This is how the camera makes the dark bits brighter and the bright bits darker, thus exposing them correctly. Don’t confuse this with black and white images, the camera does the same for ‘bright red’ and ‘dark red’. It ranks the colours on how close they are to white or black. This method of determining brightness is called luminosity. You may remember luminosity from my last blog where I touched on luminosity masks. They use the same principle to make targeted adjustments to your image.
This is how the camera determines the brightness or luminosity of colours in a scene
Anyway! You’ve just begun to understand them. But! This is where the rule can change. Have you ever tried to photograph someone wearing all black clothes and their face comes out over exposed? Even though we know the black clothes and persons face are at the same brightness, the camera thinks they’re different because it sees from black to white. As a result the camera will try to expose that black to brighten it up causing the face and other details like the sky to get blown out or clipped (too bright, we’ll touch on this shortly!). Why does the camera do this? Because the camera tries to make that black, grey. Similarly have you ever photographed a snow scene and the picture comes out very dark? That’s because the camera is confused by the white snow, it thinks the scene is very bright so it will try and darken your exposure to compensate, often underexposing the shadow areas of your image!
Thanks to Gary and Ana for a lovely day shooting their wedding!
Some examples of weird histograms. In the ring shot above if you were to just look at the histogram you would assume the image is over exposed, but it’s actually like this because the camera sees a lot of white pixels. Just like when shooting the snow scene if you were to leave the camera in automatic it would take a much darker exposure of the scene. The same can be said of the image of Gary and Ana. There is a lot of pixels at the left side of the image so you would assume it’s underexposed, but looking at the image you can see they’re both properly exposed. This is because the majority of the image is dark pixels. Garys suit and the corridor behind with no light on would all be considered “black” by the camera, therefore it would try to overexpose the bright areas to correctly expose the dark areas. To combat this you can use manual settings or the exposure compensation dial to help the camera read the scene correctly! You can learn about these things in depth on one of my lessons!
Now to discuss clipping. Clipping is when your pixels don’t fit onto your histogram. The furthest left is black, you can’t get blacker than black! The furthest right is white, you can’t get whiter than white! So if you have a huge spike touching the far left or far right of your histogram that means you have lost detail in those areas. You can see it a lot in Landscape photography, I can guarantee you have even experienced it yourself, where the sky is kinda muddy and washed out with pure white areas.
White clipping, all the pixels are pushed up to white meaning you've very little information to use in your highlights
With modern digital cameras we don’t worry as much about the blacks though. The new sensors we have pick up an incredible amount of detail in the darker shadow areas. With our whites though we need to ensure we don’t clip them. This is impossible to fix in post, you need to have that information on the histogram! Look at the ring photograph above. You’ll notice there’s a ton of pixel information to the right of the histogram, but none of it is actually touching the far right. That means none of my pixels have become “whiter than white” giving that muddy over exposed look. The best thing to aim for when shooting your scene is the have all the pixel information further to the right than the left, without touching the right. This is known as “exposing to the right” or “ETTR” for people who enjoy a good abbreviation. If you follow this rule you’ll never over expose an image or lose detail in a sky ever again! Don’t worry about the shadow areas, that detail is very easily recovered.
This is a lovely histogram showing an example of ETTR. All our information is between black and white but none of it is touching either side. It's a beatiful thing
Right. I suppose that’s it then! You’ve now mastered histograms and as a result will forever be changed as a human being. I’d love to hear feedback in the comments below, was there too much or too little info? More images or less images? Help me help me. If you think you know someone who could benefit from this blog why not share it on social media too!

Thanks for reading!

Ronan Harding Downes

Scroll to Top