Camera College: Histograms
What’s that graph shape trying to tell you about your exposures?
When you’re assessing the exposure of your photos, what you see on your camera’s screen or electronic viewfinder (EVF) isn’t necessarily what you’ll get. The brightness of the display influences how bright an image looks – with the brightness of the ambient light that you’re viewing your pictures in also having an effect. You may find that your photos are actually underexposed or overexposed when you later check them on a computer.
This is where the histogram comes in. This small graph, which can be viewed alongside your images, allows you to quickly make a more accurate appraisal of the exposure. We introduced the histogram in last issue’s guide to metering, but we’re going into more depth here, as being able to understand what the histogram is telling you is one of the fundamental skills of mastering digital photography.
You can check the histogram when you play back an image, or in real time when you use the Live View feed on your camera’s rear screen or EVF. A live histogram means you can check the exposure and make any adjustments before you take a photo.
Both of these brightness histograms do the same thing: they show you the ‘tonal range’ of an image – from black on the left through to white on
the right, with a spread of grey tones in between. The height of the histogram shows you how many pixels have registered at each level of brightness, and the histogram’s shape and position on the graph helps you to gauge whether or not your image has been properly exposed.
There isn’t really a ‘perfect’ histogram shape. The spread of tones changes according to the scene or subject that you’re photographing and the exposure settings on the camera. You would expect the histogram for a predominantly bright or pale scene to have a histogram that peaks towards the right side, while the histogram for a very dark subject should be closer to the left side of the graph. The majority of scenes, however, are a mixture of shadows, mid-tones and highlights, and are more likely to show a histogram that rises and falls across the width of the graph.
If the spread of the histogram doesn’t match the brightness of the scene or subject you’re photographing, then the exposure may be incorrect and you may want to take steps to fix it. For example, if you take a picture of a snow-covered landscape and the histogram isn’t towards the right side of the graph, the snow will appear grey in the photo rather than white. You can use exposure compensation to rectify an incorrect exposure, with positive compensation pushing the histogram towards the right (bright) side of the
There isn’t really a ‘perfect’ histogram shape. The spread of tones changes according to the scene or subject that you’re photographing and the exposure settings on the camera
graph, and negative compensation pulling it towards the left.
The histogram on the camera is small, so it can be difficult to see when only a few hundred or so pixels are overexposed; and it’s highly unlikely that the highlight alert or zebra stripes feature available on some cameras will indicate smaller overexposed areas. If recording delicate details in bright areas is important to the picture, reduce the exposure to preserve the highlights, as it’s easier to brighten darker parts of an image when you edit your images than it is to try to claw back detail in overexposed ones.
While it might seem sensible to routinely underexpose shots in order to avoid clipping the highlights, if you then try to brighten up a dark shot in photo-editing software, you can increase the amount of noise. In fact, if you shoot raw files there’s an argument for routinely overexposing your shots so that the histogram is close to the right edge without being clipped, then reducing the exposure back to the correct level when you process the images. The theory is that more tonal information and detail is recorded in the right side of the graph, and you also reduce the chance of noise in the shadows.
This technique shouldn’t be used for JPEG images though, and it’s best to aim for a spot-on exposure in the camera. Good job there’s a histogram to help you…
The histogram on the camera is small, so it can be difficult to see when only a few hundred or so pixels are overexposed