Many photographers heard about the histogram, but don’t entirely know how it works. This is the reason I wrote a lesson about it for the course Ditch the automatic mode. It’s an incredibly valuable tool and I know many photographers will have an advantage if they learn how to read it.
During the creation of the lesson I sometimes read a bit on the internet to check if I missed any handy information, but when I was writing this lesson my jaw dropped! There is so much misinformation to be read all over the internet! It was horrifying!
A few myths I encountered:
- The graph cannot toch the side AT ALL
- It’s wrong when your graph is flat on one side
- In the snow it’s ok when the graph touches the side, snow is light
- A graph should ALWAYS be shaped as a mountain
- A graph should NEVER contain of more than one peak
This list can go on and on. Within this blogpost I will explain how you understand and read your histogram correctly.
How does the histogram work?
Firstly, let’s look at a simple version of the histogram. The “photograph” below is made of a few pixels. On image 1, you can see the photograph as you would see it. In image 2 the photograph is transferred to black and white and the pixels are being counted. In image 3 you see the graphical representation of this “photograph” in the form of a simple histogram. In this case, there are three different tones: white for the background and eye, black for the bird and orange/grey for the beak of the bird. You can make three bars. The more tones of grey you have, the more bars you can draw. In reality, your histogram has a bar that goes from 0 (pure white) to 255 (pure black) and makes a graph that consists of 256 bars. This makes the histogram a factual representation of your photograph. There is no “good” or “bad” histogram, you can only use it as an aid and only you can judge if it looks like you have in mind.
To make more clear how the histogram works and what you can do with it I will explain a few different situations which you could encounter while working with a histogram.
The graph leans against the right side
When the graph leans against the right side, you have pure white pixels within your photograph. When the graph leans to this side you call it “white clipping”. By only looking at the histogram you can’t yet say if this was intentional or “wrong”. Only when you see the photograph you can judge whether these pure white pixels are ok or not.
Image 4 is a typical example of a photograph where the graph leans to the right side but the photograph itself is not overexposed. This photograph was created in the studio with a white background, the goal here was to get the graph to lean right, so the background actually is purely white.
For image 5 it’s a different story, this is a photograph of a snowy landscape. With this picture, you want to save as much detail as possible. When the graph touches the right side, you’ll know you’ve lost detail there in the lightest parts. If you see this happening when you are photographing, you can change your exposure to make it a bit darker. When you are photographing in manual mode, you can close down your aperture, shorten your shutter speed or lower your ISO. Are you photographing in an automatic or semi-automatic mode? Then you’ll need to use the exposure compensation and put it on a minus-number.
Graph leans against the left side
When the graph leans against the left side, it means you have pure black pixels in your photograph. When the graph leans to the left this is called “black clipping”. As explained in the previous paragraph, sometimes this is the intention and sometimes it isn’t. Image 6, of the horse, has been made with a pure black background on purpose. With this kind of photograph, you try to get the graph to lean against the left side. The goal is to remove all detail from the black background, you want to have an even black background.
When you are making silhouette photographs like in image 7, you are trying to get the silhouettes purely black most of the time, in this case, the graph is meant to lean to the left as well.
In Image 8 it’s not the intention to have black clipping, this image is a bit underexposed. When you see this happening when you are photographing you can remedy this by adjusting your exposure a bit. When you are photographing in manual mode you can open up your aperture, lengthen your shutter speed or up your ISO. However, are you photographing in an automatic or semi-automatic mode? Then you’ll need to use the exposure compensation and put it on a positive-number.
Graph leans against both sides
This is something that could happen when you photograph a scene with a lot of contrast, a few examples: You are photographing from within a window to the outside while the sun is shining, you are photographing against the sun during sunset or you are photographing a lighthouse with heavy shadows in the middle of a summer day.
When the graph leans against both sides, there is not a lot you can do to fix this by choosing different settings. When you see this, it means the contrast within your photograph has become greater than your camera can handle. Now it’s time to make some artistic choices, do you value to keep detail in the light area’s or are the details within the shades more important to you?
In image 9, where the man is staring out of the window, the histogram can’t be “repaired” by choosing different settings so you would see all details of both the wall and the sea and sky. Within the histogram, there is not enough space to move the graph enough to prevent clipping on either side. This does not have to be a big deal, the focus needs to be on the man and the sea and sky he’s looking towards anyway, the details of the wall are not important.
The only solution to keep detail in both area’s of the photograph would be to create an HDR photograph. You would make a photograph in such a way the dark areas are not underexposed and you would make a picture in such a way the bright areas are not overexposed. After that, you would combine both photographs in software like Adobe Photoshop CC.
Img 9 – dynamic range too big- photo by Noah Silliman
When you have to choose, most photographers won’t mind less details as much as less details in the bright area’s. So if you have to choose, most of the time it would be better to underexpose.
Is it a bad thing if the graph is leaning against one or both sides?
It doesn’t have to be! It’s a good thing to know how the histogram works and it can be very convenient to check it every now and then but you need to make absolutely sure you don’t lose your creativity as a photographer by minding the technical details too much. There are photographers who over- or underexpose their photographs on purpose. For example a portrait with the sun coming in from behind or a portrait with a lot of shadows. It could be a part of your creative process to make use of the “clipping” of the white or black parts of your photograph. The histogram is nothing more than a handy tool, it shouldn’t become a goal on itself to chase the “perfect histogram”. There are a lot of photographers who have the opinion the histogram, under no circumstances should touch the sides. Then you wouldn’t even be able to create image 11 and 12 at all!
Img 11 – photo underexposed – photo by Trevor Cole
Img 10 – background underexposed – photo by Candice Picard
Display white clipping on your camera
Many cameras have the possibility to display which pixels are completely white within your photograph by flashing from white to black. Within the manual of your camera or on the internet you can find how this works for your specific camera.
When you have a Nikon camera, you should be able to view this by looking at your display and pressing the ok-button to the top or bottom until you see the highlights being displayed as a flashing item. With a Canon camera, you click the “display” or “info” button (depending on the model), until you see this display. For some Canon models, you might have to activate this mode within the menu first.
Did you find this lesson/blogpost useful? Please let me know in the comments!
Do you want to pin this blogpost to pinterest? You can use one of the images below: