In our previous post, we looked at the basics of histograms.
Here we’ll continue that discussion and look at how we can use histograms to help get proper exposure in our images.
We already talked about how a histogram is a bar chart that shows the distribution of luminance—the brightness values—of the pixels within your image.
Maybe an easier way to think about a histogram is as a visual representation of the bright and dark areas in your photo. So when you look at a histogram, you can instantly see the levels of brightness and darkness within your image by just looking at the number of bars and the heights of those bars in the different sections of the histogram:
While there is no perfect histogram, the shape of the histogram can often alert you to possible problems in your image. In particular, the histogram can signal under- or overexposure.
Under- and Overexposure
In general, the histogram for a well-exposed image will usually show values throughout the entire range of tones, from the darkest black to the brightest white. You can see this in the image and corresponding histogram below.
The bars of the histogram spread across the whole horizontal axis, with the height of those bars trailing off towards the right and left edges of the graph. This shows that the image contains the full spectrum of brightness values, including shadows and highlights.
But if the histogram shows gaps near its edges, your image may be under- or overexposed.
Consider the following example:
The image above is underexposed and you can see that from the shape of the histogram. In this case, the bars of the histogram are shifted to the left of the graph and there is a sizable gap along the right side. This means that there are no bright tones represented in the image. Increasing the exposure (possibly by using exposure compensation to override the camera’s exposure meter) would brighten the image and produce a histogram that covers a wider range of tones.
Another example is below.
This image is overexposed. The bars on the histogram are bunched up along the right part of the graph and there is a considerable gap at the far left edge. In this case, reducing the exposure, again possibly using exposure compensation, would darken the photo and allow for a wider range of tonal values.
One of the most important things to watch out for when looking at a histogram is the behavior of the graph at its farthest right and left ends. Specifically, look for a spike along the edges where the bar pattern bunches up and bleeds off the edge of the graph. This type of pattern is called clipping and it’s a sign that the image is losing detail due to exposure problems.
If the clipping occurs on the far left of the graph, details are being lost as a result of overly dark shadowy areas. You can see an example of this below.
If, on the other hand, the clipping occurs along the right edge, your image is losing details in the highlights.
So, what do you do if you see clipping in your histogram? Well, that depends.
In the two images above, the gap in the respective histograms suggests that an exposure adjustment is probably in order. When you increase exposure, the bars of the histogram shift to the right and when you decrease exposure, the bars shift to the left. So adjusting the exposure in the above photos—decreasing the exposure for the overexposed photo and increasing it for the underexposed one—shifts the bars to fill in the gaps on the edges of the histograms and allows a larger range of luminance values to be represented in the image.
But it doesn’t always work this way.
Consider the two photos below along with their corresponding histograms.
The first would seem to be underexposed from the histogram:
This second photo may at first appear to be overexposed from the histogram:
In the above images, you can see that these histograms show clipping and indicate that these is a loss of detail in the dark areas (Figure 12) or the highlights (Figure 14). But, unlike the two earlier photos, there aren’t any gaps in the histograms for these two images. That means that the entire range of tones is represented in the photos. And it also means that some of these tones would be lost in the shifting of the bars that would occur if the exposure were increased (Figure 12) or decreased (figure 14).
But more importantly, the subjects of these images—the most important part of the photograph— are properly exposed. The areas that are losing details aren’t essential to the photos. Overall the exposure for both of these images is good.
So that’s it. That’s how you can read histograms and use them to point out possible exposure problems.
Keep in mind that, with all of this, the bottom line is that the histogram is an essential tool both when you’re shooting and when you’re editing. But it’s also just a tool. While the histogram can help to point out possible exposure problems, the final picture should always be based on the type of photo that you are looking to shoot, the image that you want to create of your subject and your intended artistic effect.