One of the main advancements camera design in the early 80s was the invention of auto-exposure modes. The new invention allows the camera’s exposure meter to calculate the proper exposure for each scene without manually memorizing exposure combinations of shutter and aperture values. Virtually all in-camera meters utilize a unit called “reflective meter”, this means is the meter takes a reading of the light reflecting from the subject.

Figure 1 illustrates how a typical scene is metered by a camera. Notice that the light source provides the incident light (actual light falling onto the subject) and the subject will reflect the light in a certain manner where the camera’s exposure meter can pick up the intensity of light in order to calculate an appropriate exposure.

In most situations, a reflective light meter works well, especially when there are a lot of midtones in our scene. Reflective light meters are calibrated to a certain percentage of gray, usually between 12-18% gray. Referring to our tonality scale in Figure 2, that would equate to about the midpoint between black and white, with a rough value of 127 in our 0-255 scale. Metering for midtone is a good arbitrary range to work with as metering in the middle of the scale would allow a greater exposure flexibility on both ends of the spectrum.

While this works really well in most scenes, the design presents a problem as well, because whatever area it is metering, it will consider that area as midtone in the luminance scale. The meter cannot distinguish if the area being metered is actually dark or bright, all it can see is “that’s midtone”.

The examples below will illustrate the way our camera meters work.

Figure 3 has a scene where a white background and a black background are present. The white area is a piece of large paper, while the black area is a black fabric from my reflector. If we take a photo with only the two extremes present (see Figure 4), we can see that the camera has little problems exposing the image correctly.

The histogram in Figure 5 shows that the camera was able to capture the black as black, and white as white, with almost nothing in between.

Great results.

However, if we take the black background out of the scene and leave only the white background and take a photo, the resulting image comes out gray (Figure 6), and the histogram shows a peak near the center.

When we remove the white background and only shoot the black background, the resulting image is roughly the same. We don’t get a black image, but a gray image as well as seen in Figure 8 and 9.

What happened here? Remember that our camera’s meter can only see midtone. Anything it tries to meter will be rendered as midtone, and that’s the main shortcoming of reflective meters. If your scene is predominantly brighter or darker than midtone, the camera’s meter will not identify that difference and will render everything back to midtone.

By adding some elements in front of a predominantly white background, such as in Figure 10, the camera is easily “fooled” thinking that the white background is too bright and in turn it changes the exposure to obtain a midtone exposure. The result is an exposure where the wall became gray, and the subjects are underexposed.

Figure 12 and 13 shows the correct exposure with the large white area peaking near the right edges of our histogram.

The same thing occurs when you try to photograph a scene against a black background, as presented in Figure 14. The camera thinks it’s too dark and the meter tries to turn the black to a midtone reading. The result amounted to a gray background and overexposed subjects. You can see the histogram in Figure 15 that the peak of the histogram isn’t near the left edge, but almost in the middle.

Figures 16 and 17 shows the correct exposure against a black background. Notice the peak of the histogram is leaning towards the left of the scale and the subjects are not overexposed.

What does all this mean? Quite simple, when using your camera to meter a scene, remember that the camera is metering midtone. You have to learn to determine the brightest and darkest areas of your particular scene and find the middle-ground of the two and meter from there.

Thankfully, most modern cameras utilize a very intelligent multi-segment metering system where it dissects the scene into different smaller sections and average the readings from each of the segment to create an exposure.

Figure 18 is a simplified version of how a multi-segment meter splits the scene into 12 different areas and take individual readings on each of the square before averaging all 12 squares to come up with a meter reading. This increases metering accuracy tremendously in most cases and most modern cameras read from more than 30 areas in a scene before making an exposure.

However, it is not fool-proof, as Figures 6 to 14 shows, when a scene has too much bright or dark areas, the meter will be challenged to provide an accurate exposure.

In our next article, we will discuss how to use your camera’s metering mode options and exposure compensation to obtain the correct exposure when auto-metering fails.

About the Author