The Histogram

It can be difficult to accurately judge whether a photograph is correctly exposed or not. This is not made any easier by the LCD screens that feature on DSLR cameras – these often do not show the image very accurately or in very good quality. Luckily there is a tool on most DSLR cameras that can help with this. This is the histogram, a type of simple graph that quickly shows you whether a photograph is over or under exposed. The histogram can usually be seen once the photograph has been taken, when using the camera’s review function.

The basic idea of a histogram is to show the brightness of every pixel in the photograph. Every pixel is assigned a number between 0 and 255 representing its brightness, with 0 representing the darkest pixels possible and 255 representing the brightest pixels. The camera counts how many pixels have each number, and then produces a chart similar to the one shown below, with the number of very dark pixels on the left and the number of very bright pixels on the right. In between you can see the number of pixels at each intermediate brightness.

Notice the contrasting brightness between the sky and ground.

The contrasts are visible in this histogram, with the darker ground on the left and the brighter sky on the right. A small amount of saturation is also seen on the far right side of the histogram.

Take a closer look at the photograph and histogram shown above.  You can immediately see that there are two regions in the histogram – a kind of jagged peak on the right and a smoother hill on the left.  Looking at the photograph it’s quite easy to see what these correspond to – the small hill on the left corresponds to the darker ground while the peak on the right corresponds to the brighter sky.  At the very far right of the histogram there is a sudden peak corresponding to the very brightest pixels.  This indicates that there is some saturation in the photograph, and so some details have been lost. In this case the saturation is quite mild, and did not badly affect the photograph.

The histogram gives us a nice way to think about exposure.  If we increase the exposure this will have the effect of shifting the whole histogram to the right.  Likewise decreasing the exposure will shift the whole histogram to the left. Let’s take a look at the histogram for the same picture, but with the exposure increased by 0.3 stops. Now we see the far right side of the histogram is much more dominant.  This is a good indication that more pixels of the photograph are saturating and that it would be a good idea to reduce exposure.

The peak on the far right side shows a large increase in saturation when the brightness is increased slightly.

As a general rule you should aim to reduce lost details by adjusting exposure to avoid a histogram that is too far to the left or right.  Your camera will automatically try to do this unless you use M mode.  That said, there are times when you will want to have a histogram that is concentrated either to the right or left.  This depends on the scene you are photographing, on the natural light levels and on what you want to do with the photograph.  At this stage it’s helpful to build experience with both exposure and the histogram by getting into the habit of checking the histogram for every photograph you take. It’s also useful to try playing around with the exposure to see how this changes both the image and the resulting histogram.  In general you should aim to get the exposure correct at the moment you capture the photograph.  If in doubt, it’s better to slightly underexpose rather than overexpose, as underexposure is easier to fix in post processing.

Two keyholes in an old door

The two keyholes are very dark while the rest of the image is evenly lit

The even lighting is clear in this smooth histogram, with the small peak on the left side representing the dark keyholes

The image above shows a photograph with a very central histogram.  The small peak on the left of the histogram corresponds to the dark (almost black) key holes, while the smooth main peak makes up the rest of the image.  This type of histogram is typical of images with even lighting over the whole photograph, and does not lean towards being bright or dark.

A dark but low contrast image of a turtle swimming in an aquarium

The narrow, smooth peaks indicates that this is a low contrast and evenly lit photograph

This image is much darker.  We can refer to this image as being “low key”, as most of the histogram is concentrated on the left.  Take a look at the width of the histogram of both this image and the previous one.  Both histograms have a similar shape, with a smooth peak trailing off to the sides.  However the second photograph has a much narrower peak than the first.  This gives us a measure of the contrast present in the image, that is, the difference in brightness between the darkest and brightest areas of the photograph.  We can say the second image displays low contrast (as there is not much difference in the brightness), while the first image shows a high contrast.

The final topic we will discuss today is dynamic range.  Dynamic range is the maximum difference between the darkest and brightest pixels that your camera can show.  If we look again at the first histogram we saw in this lesson, we can see that there are two peaks that fall comfortably within the histogram.  In this case the dynamic range of the camera was sufficient to take the photograph without too much loss of detail.

The bright light from the church windows contrasts sharply with the dark church interior. The dynamic range of the camera struggles with the high contrast and forces the photographer to choose between the windows or the church interior.

The histogram shows the problem clearly, as we see the two peaks are too far apart for the camera to capture. In this case techniques like HDR may be used to capture both the church interior and the bright windows.

But we can imagine another photograph where there is a bigger contrast between the light and dark areas, as in the photograph of the church window above.  In this case the two peaks have moved apart, and we find that the dynamic range is not sufficient to capture the details of both the dark interior and the brightness of the light shining through the window.  One way to overcome this is to sacrifice detail in one area of the photograph, or to alter the composition so that the contrast is no longer so high.  An alternative technique is to use High Dynamic Range (HDR) photography. In this technique two photographs are taken with different exposures in order to capture the bright and dark areas separately. These two images are later combined using software to produce a single high contrast photograph.

No comments yet