Understanding the histogram

The histogram is a graph that represents the tones of all the pixels in an image. The left side of the graph represents pure black and the right side pure white. Everything in the middle are mid-tones; the very middle being 18% grey. The higher the peak, the more pixels there are of that tone. There are 256 brightness levels in a histogram. On the far left, pure black is 0 and pure white is 255.

What should the histogram look like?

In a perfect world the histogram would look something like the image to the right. Ideally you don’t want to have any large gaps on either end of the graph. example of a perfect histogramThe tones should just touch the end on both sides and rise to a peak in the middle of the graph somewhere.

What information does the histogram tell us?

histogram of overexposed image

Histogram for an image that is overexposed

Whenever there are tones pushed up against the end of the histogram, you have a problem. When tones get pushed to black or white they begin to lose detail (or clipping). Detail that is lost from clipping cannot be recovered. Even post-processing won’t bring it back. Once the detail is gone, it’s gone, but it’s easy to fix. The exposure needs to be adjusted and then recomposed again. The other big indicator to look out for is when there are empty gaps on the end of the histogram.

The histogram for real-life images will never look like a perfect bell shape. Every image has a different combination of tones in it. High-key images have a lot more detail and pixels on the right end of the histogram than they do on the left. This is perfectly acceptable as long as the detail doesn’t push into the end of the graph. Below are several examples of images and the histogram that accompanies them. They will help you become familiar with what to look for.

Histogram Examples