How to read an RGB histogram

One of the best ways to determine if your image is exposed the way you want it is to check the histogram. However, when you start talking histograms, a lot of people get that glazed over look in their eyes. It’s a fancy word for something that’s really quite simple. Knowing what the graph means, and how to read it, are important things to understand however.

You may have seen these in programs you use, or on the back of your camera. They are histograms, a visual representation of data in your image. A histogram is a bar graph that shows you the distribution of luminance values in a digital photograph.

The horizontal scale of the graph is broken into 256 increments showing you tonal values, from black (0) on the left, to white (256) on the right. The vertical scale represents quantity of data at that increment, and is dependent on a number of factors including bit depth, color space, and file compression (if any). Generally speaking, though – the higher the peak at that tonal level, the more data is available there. A low key image, one that is generally underexposed with lots of dark information, would be heavy on the left side. Highlights and bright areas show up as spikes on the right side.Furthermore, histograms can be broken down into a master channel, and into the various color channels (Red, Green, Blue). Here’s an example of two shots – one dark, and one bright – in Adobe Lightroom. Notice where the most intense peaks fall in both images.

Notice the difference in the color of the histograms as well. Lightroom (as in Photoshop when you have your histogram in the expanded view) will show you not only a master histogram (usually depicted as grey or white) but also the colored channels, giving you an idea of the amount of luminance information in those particular channels. In the first picture of the candle, there’s a lot of yellow and red, but no blue. In the second picture, there’s a lot of blue and yellow, but not so much red. Bear in mind, these histograms are made up of three layers of color that represent the Red, Green, and Blue color channels. Gray appears when all three channels overlap. Yellow, magenta, and cyan appear when two of the other RGB channels overlap. Yellow is made up of the Red + Green channels, Magenta are the Red + Blue channels, and Cyan equals the Green + Blue channels.

This is handy information to know, and Lightroom will go so far as to let you drag parts of the histogram to make changes. They’ll even break down the horizontal scale for you in terms that photographers can easily understand… so you can see whether you’re dealing with blacks, fill, exposure, or recoverable highlight information.

As mentioned above, this information is usually also available on your camera. Some cameras will show a simple composite histogram, and newer or more advanced cameras will let you break it down so you can see the Red, Green, or Blue channels separately. The rule of thumb is to expose for the highlights. Highlight information, on the right, can easily be clipped by the method in which a digital camera meters a scene. If it is, you’ve likely lost that information, so underexposing a small amount is preferred.

This is most important if you’re shooting something that’s intended to be white – like snow, or a white wedding gown. The way a camera meters is by trying its best to balance the luminance levels in a scene. If you underexpose by a third or half stop, you can then preserve those brighter details and using software, you can bright them up to an acceptable level manually.