Histograms

Histograms are a very valuable tool in evaluating digital images.  What is a histogram?  It is a graph that looks like this:

Shadows         Midtones         Highlights

In this type of graph, the left side represents pure black, the right side represents pure white, and in between represents all the shades between black and white.  The height of each area in the graph represents how much of each shade is present in the image.  It may be helpful to think of the graph like this:

Black   -   Shades of Grey   -   White

Note that histograms are not just for black and white images.  They can represent the tonal range of images of any colour.  Image editing programs such as Adobe Photoshop usually have a Histogram feature.  Many digital cameras have this feature as well, and that is what we want to discuss.

What's the Use?

How can histograms help us to get better photos?  They can show us if a photo is incorrectly exposed, without having to trust the LCD screen on the back of the camera.  If you've ever looked at a digital image on a computer and said, "Hey, it didn't look that dark in the camera", you know what we're talking about.

Take a look at the histograms below, and the images that correspond to them.  You will notice that the underexposed image's histogram is weighted to the left, and the overexposed image's histogram goes off the end of the graph to the right.

Fig. 1 - Underexposed               Fig. 2 - Correct Exposure               Fig. 3 - Overexposed

The correctly exposed image (Fig. 2) shows a nice smooth "hill" that begins and ends inside the graph.

The underexposed image (Fig. 1) has less information on the right side of the graph, where the highlight detail should be.  Notice that the "hill" actually starts somewhere off to the left of the graph.  This is an indication of "clipping", which means some of the information that should be in the picture is not.  Our nice hill has been clipped!

The overexposed image (Fig. 3) has a graph that is skewed to the right, and displays clipping in the highlight range.  This is obvious when looking at the baby's cheeks in the image; they are "blown out".  In digital photography, clipped highlights is probably the hardest problem to fix.  You are better to err on the side of underexposing (stacking the histogram toward the left of the graph) than overexposing.

Be careful not to only rely on the histogram to judge your exposures.  Many scenes should have a skewed histogram, because the scenes themselves are weighted toward a specific tonal range.  Look at the images below:

The above images are all exposed correctly, but because of their subject matter they have some funny looking histograms.  With some practice, you can become quite adept at prediciting what an image histogram should look like based on a correct exposure of the subject matter.

If your camera has a histogram function (see Fig. 4), knowing how to read the data can be very helpful in determining if your exposure is correct.

Fig. 4 - In-camera histogram display
Image courtesy of www.dpreview.com

Note the blinking areas in the above display.  The areas that turn from white to black are an indication of clipped highlights.  Remember, clipped highlights are something we do not want.

Of course, you can't retake every incorrectly exposed image.  Sometimes the moment is gone before you can say "histogram".  But for those situations where you can review the histogram on the back of your camera, you will be better able to judge if you nailed the exposure on the first try.

This article is © Copyright Photo Express foto source