How To Read A Histogram
How to read a histogram. A histogram is a graph showing the spread of tones in a photograph.
The darkest tones are on the left of the graph, with the brightest on the right.
In this article I hope to explain what it is and how to read a histogram.
There are 256 points along the width of the histogram.
They start at black numbered 0 at the left edge and continue to point f white at the right.
The graph show the tones in the image, not the colour. Where there is more of a particular tone, the graph is higher.
Most digital cameras have the facility to show a histogram when you’e composing a shot.
They’ll also be able to show it when you’re reviewing a photograph that you’ve taken.
Different cameras will have different settings that show the histogram.
If you check your camera’s manual you’ll find how to display it.
Most imaging software will also be able to display a histogram.
Sometimes it’s available as one block representing the aggregate of all colours.
Other applications will also show the red, green and blue values separately.
They look intimidating at first but it’s easy to get to know what they’re all about.
The best way to explain things is with pictures.
In this case I’ll show some photographs with their histograms and explain what’s going on.
How a histogram reads a night time photograph
The first photograph I have is a night scene.
So let’s see how to read a histogram.
This shows a hotel and it’s reflection in a lake.
The hotel is situated behind the lake.
It’s was night time when this photograph was taken.
As you can see it’s a very dark night.
The hotel is well lit, and the wall facing the camera is quite bright.
There are some lights shining straight at the camera which seem to be completely white.
The reflections in the water are mostly lights.
Despite this the photograph is very dark overall, with most of it black, or near black.
The histogram is pushed very much to the left.
When the graph reaches right up to the top it shows that there’s a lot of a particular tone.
In this instance the graph touches the top for a large portion at the left, or black zone.
That shows how dark the photo is.
When the graph touches either the left or right end it shows a lack of detail.
At the left end, the dark at the very edge is totally black.
If there was a similar happening at the other end there would be total white.
Away from the edges a tone can be lightened or darkened by software adjustments.
Pure white will not respond to these adjustments.
Pure black might show some poor quality detail.
It’s better for editing it there’s more very dark areas than very light ones.
Avoiding tonal extremes
It’s not desirable to have any part of a photo that lacks detail, or is blown out.
For that reason it can be helpful to check the histogram before shooting.
If the histogram touches the sides, settings can be changed to avoid blow out parts.
There’s also a setting on a camera which will show that blow out is about to happen.
This has to be located in the menu and turned on also.
Again refer to your camera’s manual to see how.
When it is switched on, it will flash red where there’s pure white, and blue where it’s pure black.
Most image editing software also has the facility to show blow out at either end.
Once more white areas will be shown os red and black areas as blue.
While editing photos you will probably be changing things like brightness, contrast, hue etc., and you’ll see where to stop.
When you see the red or blue appearing over the white and black, it’s time to go back a little.
Back to our first histogram, it does reach to the right edge, but faintly.
The lights ensure that there is a spread of tones, but the graph trickles towards the right.
When you know how to read a histogram you’ll see this in the graph.
How to read a histogram of a snow scene
For the next image I go to the other end of the spectrum, with a snow scene.
There seems to be as much white in this shot as black in the previous one, it’s not as pure though.
Actually it doesn’t seem to extend all the way to the right.
There’s a lot of the graph towards the right, and it goes right to the top.
This shows that there’s a concentration of light but not white tones.
The snow at the front of the photo is in the shade a little darker, and shows a little grey.
The sky at the back has a definite little hint of blue in it also.
The field in the middle looks very white though.
However the histogram doesn’t seem to see any white at all.
There’s trees and bushes right throughout the image which look quite black in places.
According to the histogram there isn’t any pure black in them.
There’s a few spots in the foreground which look almost black, but the histogram doesn’t think so.
This looks good for a snow scene.
The big area of snow looks very white, as it should.
The area at the front is flanked by a tree to the left, and a house to the right.
One of these is the cause of the slight shaded area.
What the ideal histogram should look like
The ideal histogram should rise slowly from either end and peak in the middle.
That’s regarded as a correct exposure.
This next shot seems to have an almost ideal graph except for the blown out sky.
Actually I think it’s quite a drab shot even when disregarding the sky.
When you’ve got a blanket white cloud across the sky it can turn out like this.
Off course the cloud wasn’t totally without some colour.
It could have showed some by using an ND filter or some exposure compensation.
But according to the histogram it’s totally white.
The main colour in this photo is green.
There’s quite a lot of shades of the colour, but none that are very strong.
Some of the trees do look like they’re a dark shade of green, but the histogram doesn’t show that.
The graph doesn’t rise very far except at the white end.
In software a boost to the contrast would have stretched the graph closer to the ends.
It wouldn’t do anything for the sky, but it would liven the rest of the image.
Mainly mid tones
The tones in the next photo are somewhat similar.
There’s no sky in this photo, with or without colour showing in it.
It’s dominated by shades of green, as the previous photo was.
This time the graph does rise high in the dark half.
I’ve presented this photo straight from the camera.
Normally I would boost the contrast at the very least.
This would obviously change the graph, pulling it towards the sides and increasing the tonal range.
I want to show how to read a histogram before editing it so you know how to take charge.
Not knowing how to read a histogram may hamper that.
Taking a photograph is different to painting a picture.
While you can change the settings on a camera, you have to shoot what’s in front of it.
Some subjects, such as this one, don’t have a full tonal range.
Paint a picture you can use any colours or tones you like.
A slightly deceptive histogram
The fifth photograph, the fruit and Easter egg has a histogram that stretches almost from end to end.
Most of the weight is in the darker half with the light half barely showing.
To me the photo looks lighter than what the graph seems to indicate.
The main tones are in the dark end of the range.
There are small highlights on the fruit and wrapping paper, and light tones in the table top.
This stretches the graph over to the right, if faintly.
While the tone is mainly dark it doesn’t stretch to the left.
Different looking histograms
Although histograms give the same information, different software programmes show them differently.
The histogram attached to the last three photographs were generated by a different software to the others.
It’s good to get used to looking at a variety of different looking histograms.
They all tell the same story and you’ll soon get used to reading it regardless of how it looks.
How important is a histogram
It’s important to know how to read a histogram and what it does.
It can tell you how well your photo is or isn’t exposed.
But do you need a histogram to tell you that.
Some subjects are on the dark side and some are bright.
The histograms for these subjects should show that.
If you try to fix the graph so that it reads like a perfect exposure, you may ruin your photograph.
Where you’re photo is very obviously incorrectly exposed surely you’re going to know when you look at it.
If it’s slightly out, does it matter very much.
The look of a photograph is a personal thing.
Regardless of what a histogram has to say, different people will like a different look.
I don’t always look at the histogram when I take a photograph.
Personally I edit it to look the way I want.
Rarely if ever will I fix it the way the graph thinks I should.
I think that it’s better to edit a photo to look the way you like it.
Whether you take notice of it or not it’s important to know how to read a histogram.
If you decide to ignore it and please yourself that’s your decision.
Still when you know how to read a histogram you do so by choice not by accident.