Looking at Histograms

A histogram is a graph that represents the tonal range of a digital image, going from shadows on the left to mid-tones in the middle, and highlights on the right. The data reflects the tonal range of the image you have open at the time. Every single pixel shows on the graph.

For the most part, there is no “right” or “wrong” histogram in digital image editing, but of course a histogram does show up any problems with an image. For instance, if a photo is underexposed, the data will bank up on the extreme left of the histogram. Overexposed detail will bank up on the extreme right. Other problems such as colour casts will also show up on histograms.

Different Types

There are several types of histogram that you’re likely to encounter. The “luminosity” or “luminance” histogram concerns itself purely with the black and white brightness of an image and whether or not there are any areas of complete underexposure or overexposure. What it doesn’t do is break down the data into the individual red, green and blue primary colors which combine to make light. That is a bit of a shortcoming when editing colour digital images.

Another type of histogram is the “RGB” histogram. This tells you more about whether or not red, green and/or blue (RGB) colors are correctly exposed. In that respect, it’s more useful than a luminosity histogram, but it’s still a little crude in that it only displays this colour data collectively. (By the way, when colour is overexposed or underexposed—aka “clipped”—it turns into a blob of colour without any textural detail. That effect is known as “posterisation”.)

Probably, the most useful histogram you’ll use in digital imaging is the “colors” histogram, as it’s currently called in various forms of Photoshop. Adobe Lightroom also has this type of histogram. It displays all three RGB colour channels individually, and gives luminance info at the same time. Thus, you can see everything you need to see and address any problems accordingly.