How to use the histogram displays in your camera to improve your photos.

All digital cameras are able to display a histogram on their LCD monitors and/or in their electronic viewfinders. It’s one of the benefits of digital capture, yet it’s often ignored because many people don’t know how histogram data should be used.

Most casual snapshooters check the image reproduced on the monitor screen after a shot is taken – the slang term for which is ‘chimping’. But while chimping lets you check sharpness, depth of field and, to some degree, colour reproduction, it’s less accurate for checking actual exposure levels due to the influence of ambient light hitting the screen.

The displayed image on a monitor screen will look different in bright and dull conditions, indoors and outdoors and under different types of lighting. Screen brightness and other factors can also affect the displayed image. EVFs overcome most of these problems by excluding most (if not all) of the extraneous light and even when images are viewed through the viewfinder, histograms play a useful role in determining whether a shot is correctly exposed.

Reading a histogram

Brightness histograms are found in even the most basic cameras (including on cameraphones). The display is very simple: the horizontal axis plots luminance (brightness) values from pure black at the left end to pure white on the right, while the vertical axis plots the number of pixels at each step along the horizontal axis, which contains 256 discrete steps with a JPEG image.

To read a histogram you must be able to visualise how the various tones in a scene will be fitted onto the graph. Where will the brightest and darkest tones fall? How will the intermediate tones be distributed?

The histogram in the lower left corner of this image is the same as used for the diagram above. Note the lack of any true black or even deep shadows in the image and the matching low density of white areas. Most of the tones are on the brighter side of the mid-point along the horizontal axis, reflecting the slightly higher-key tonality and relatively low contrast in the image.

The histogram for a scene in which dark tones predominate. See how the graph pushes up against the left hand end of the horizontal axis. The narrow band of bright tones shows that a limited dynamic range has been recorded.

Clipping occurs when no details have been recorded in either highlights or shadows (or both).  Blown out highlights are a common indication of highlight clipping, while blocked-up shadows characterise shadow clipping. In the dark scene shown below, the histogram indicates there is too little information in roughly 70% of the image for details to be recorded.

This smartphone photo shows both highlight clipping (circled in red) and shadow clipping (circled in green).Even though the histogram shows a good range of midtones have been recorded, no detail was captured in the clipped areas.

Shadow clipping is usually more noticeable than highlight clipping, since the latter can involve fewer pixels. A narrow spike on the right edge of the histogram is the surest way to pick up clipped highlights in an otherwise correctly exposed image.

Clipped highlights needn’t necessarily be a problem, particularly in snowy scenes, like the one below, where people don’t expect a lot of detail to be recorded in the brightest areas.

Clipped highlights are less problematic in scenes containing areas of snow.

Many cameras include flashing indicators or ‘blinkies’ that will alert you when clipping is occurring so you can adjust the exposure accordingly to prevent (or at least, minimise) clipping. Unfortunately, even with the latest image sensors, there will be times when the brightness range in a subject will exceed the dynamic range the sensor can record. That’s where shooting raw files can be an advantage, because they have a greater dynamic range than JPEGs.

For JPEGs, 8-bit sampling will encompass 255 brightness levels. If you’re shooting raw files, sampling depth will be between 12 (4096 levels) and 14 bits (16,384 levels). With higher bit depth, more of the captured dynamic range is retained, making it easier to access tiny differences in brightness levels when images are edited.

It’s worth noting that both highlight and shadow alerts are usually based upon the way JPEGs will be reproduced. They won’t show details that could be reproducible through the wider dynamic range in raw files. There are also instances, notably with grossly over-exposed subjects, where blinkies don’t display at all.

RGB histograms

As well as brightness histograms, more sophisticated cameras can display additional histograms for each of the three primary colours: red, green and blue. Colour histograms show the brightness distribution for each colour individually, letting you see how different colours are distributed on the tonal scale.

Most RGB histogram displays include a brightness histogram.

They can be used to determine whether each colour is correctly exposed and will provide an indication of the overall colour balance in the scene. They will also show if one colour is overexposed and clipped, something that isn’t possible with a brightness histogram.

The most common use of colour histograms is to determine whether the white balance in your image is correct. If the graph peaks in the same place on the histogram for each colour channel, the colours in the image are balanced and the white balance will be neutral.

When they don’t, the highlights will contain a colour bias and if one graph goes off the end of the scale, that colour will be clipped. Areas where significant colour clipping occurs will lose all texture associated with that particular colour, although they may retain some brightness texture if the other two colours haven’t been clipped.

Also see:
How histograms can help with tricky subjects
Expose to the Right Technique (ETTR)

Article by Margaret Brown – see Margaret’s photography pocket guides  

Excerpt from  Photo Review Issue 78   

Subscribe to Photo Review magazine