Hone your exposure control through an understanding of histograms and light meter readings.
What Is a Histogram?
A histogram is a bar graph that was originally created for statistical analysis. As digital photography was developed, designers incorporated histograms into cameras to provide a graphic representation of the digitally recorded image. However, even today their function is not fully understood and the information they provide is often incorrectly used.
Figure 1. Anatomy of a photographic histogram.
*Frequency indicates the number of each tonality occurrence.
As shown in Figure 1 above, your digital camera creates a histogram that describes the tonal range of each scene you capture. The camera's processor locates each picture element (pixel) on this horizontal scale according to its relative brightness from 0 (black) to 255 (white). Pixels of equal brightness are stacked vertically one upon another to create lines of varying heights. The result is a graph of very fine vertical lines which can appear as a smooth curve, a series of jagged lines, or a combination of both.
A histogram cannot tell you about lighting ratios, the best flash-ambient mix, or if the subject is properly exposed. That is why using a light meter, along with your camera's histogram, is the best way to assure optimum results that can be easily reproduced.
Digging Deeper into Histograms
It is not always easy to make sense of a histogram showing the full tonality of a scene, so we have created the graphic examples below to help simplify the process. In Figures 2-6, the tonal steps are arranged starting with no midtones, simply black and white. Each successive figure adds midtone steps. The steps in each figure are of equal size and shape. The resulting histograms clearly show each tone as an equal height line that is uniformly distributed across the horizontal scale.
As Figure 6 illustrates, even when the number of steps is increased to a degree at which the impression is a continuous gradation, the resulting histogram appears smoother but can still show all the discreet tones.
Photography would be much easier if histograms had instantly recognizable good shapes and bad shapes and if there were "ideal shapes" that guaranteed perfect results. The truth is, they are just graphs. All of the histograms in this article accurately describe the pictures located near them. Their shapes are all very different because the tonal distribution of the images they represent is also very different. At best, a histogram is a guide; it is up to you to determine how to use it.
A histogram offers a great way to judge whether the brightness range of a scene will fit within the dynamic range of your camera. That is, if it is distributed fairly evenly within the confines of the left and right sides, you will most likely have an easily printable image. However, if the shape of the histogram is pushed up against either side, you have to understand the difference between normally exposed, high-key, and low-key images, and the possibility of losing highlight or shadow detail in the scene you are trying to record.
Histograms are not typically helpful in determining the exact exposure of the subject, the effect of lighting on the subject, or variations in light mixtures. And as histograms appear in a small window on the back of your digital camera, a quick glance is often not enough to make a quality determination.
Histograms and Light Meters: Four Ways They Compare
Although everyone has their own preferences and shooting styles, photographic imaging can be optimized by understanding what a light meter and histogram can show you in the four scenarios presented here. Virtually every common shooting situation is represented, and each example shows proper exposure as determined by a light meter while including an interpretation of the accompanying histogram.
(1) Avoid subject failure and expose for proper high-key and low-key scenes
The most common problem of built-in camera meters is subject failure. It caused the exposure system to try to "average" the reflected value of the tones of these two scenes, Figures 7 and 8. The left halves show that the results of accepting the camera meter settings would have been underexposure (trying to make the white plate middle gray) and overexposure (trying to make the black plate middle gray). Proper exposure (right halves) was achieved by using a handheld incident meter, which measured the light falling on the subjects and determined that both subjects required the same exposure.
Figure 7a (left) Figure 7b (right)
Figure 8a (left) Figure 8b (right)
Figure 7a high key (camera)
Figure 8a low key (camera)
Figure 7b high key (incident meter)
Figure 8b low key (incident meter)
The histograms, while inadequate for determining proper exposure, show the bias in tonal distribution and give a good indication of how high-key and low-key subjects will reproduce.
(2) Controlling multiple light sources, setting ratios, and adjusting tonal range
When working with more than one light, a meter is essential to evaluate and compare each light source to determine both proper exposure and the effect that each light will have on all parts of the scene, as in Figures 10-15.
The histogram is useful to gain an understanding of how the overall tonal range will reproduce and to alert you when extreme contrasts (pixel blocking on the right or left side) could cause a loss of image detail. However, as it is very difficult to interpret the visual effect by observing the respective curves, the histogram is not a suitable guide for positioning and controlling the power of each light source—especially when making small 1/10th f-stop adjustments.
(3) Balance ambient light and flash on location
Even with the best TTL systems, proper rendering of flash and ambient exposure can be unpredictable, especially when subjects are not "average" and backgrounds are bright or non-reflective. A handheld meter ensures accuracy by separately measuring both kinds of lighting.
Histograms do provide an interesting view of the distribution of tones, and indicate difference in the effective contrast of the lighting. However, there is very little useful information about how the ambient light and flash relate to each other in terms of brightness (ratio) or optimum, combined exposure.
(4) Avoid under- and over-exposure
Photographers who learned their craft by underexposing slide film to pump up the saturation should pay attention to Figures 19-23, which show only subtle changes in the histogram when compared to the metered f-stop value and resulting image. Underexposure equals image loss. Relying on a handheld meter, instead of the histogram, ensures accurate, repeatable exposures.
Photographers who learned on negative film, routinely overexposed, and let the lab compensate should pay attention to Figures 23-27, which show only subtle changes compared to the metered f-stop value and images. Overexposure equals image loss. Relying on a handheld meter instead of the histogram ensures accurate, repeatable exposures.
These photographic results show how even a half-stop variation in exposure can cause a rapid blocking of tonal information. Histograms show this in a much more subtle way, but it is difficult to impossible to see them on the average DSLR panel. And as there is no detailed horizontal scale (even in most software; see Figure 28), it's hard to visualize and gauge that the error in each exposure here equals a half stop. So there's almost no chance you'll be able to use a histogram to fine-tune settings to get within the ideal 1/10th of an f-stop.
Figure 28 RAW processing window
Film and pixel quality still suffers.
In today's digital age, the technology involved in photography is much more advanced. However, professional-quality results still require the best techniques and tools.
Digital capture has a dynamic range and latitude roughly comparable to transparency film. If you've built the skill to work within these limits, you're off to a good start. But if you're accustomed to the wider latitude (extra margin of exposure error) that color negative film affords, it's time to change your thinking.
It may seem that any miracle can be made to happen in the digital darkroom, but correcting a bad exposure has its limits. What may first look like a simple brightness correction can result in a loss of image detail that cannot be "fixed."
Figure 29 in-camera meter exposure
Figure 30 digitally pushed correction
Figure 31 handheld meter exposure
Figure 32 no compensation required
In Figure 29, the exposure was calculated by the camera's built-in system. It "observed" the excessive amount of light reflected off chrome and white background and calculated an exposure to maintain detail in these areas. Averaging in the greater brightness caused underexposure of the overall image (subject failure). Figure 30 shows the result after it was "fixed" in the computer. For Figure 31, the Sekonic L-758DR meter was used in incident mode to determine proper exposure. The result (Figure 32) looks great without any computer manipulation. At these small print sizes, the image quality of Figures 30 and 32 might seem equally acceptable. However, when enlarged and inspected (Figure 33), it's clear that the quality of the "fixed" image is much lower. By "push processing" the image in postproduction, the software "stretched" the limited information and could not fill in or fake the missing data. This literally caused gaps in the range of tones as shown in the histogram.
Of course, higher-end digital cameras and digital backs will yield better results with some variations in proper exposure. And multiple-exposure techniques as well as RAW capture will allow one to effectively extend the exposure latitude of the image. However, once again, these techniques will add additional computer processing time and require a good knowledge of postproduction software techniques.
Now, remember why you first started taking pictures. What are the keys that will get you more time shooting? A histogram and a light meter.
Do You Need a Light Meter?
Yes! While histograms provide a convenient way to see how an entire scene will be reproduced (or not), they really don't tell the whole story. They can show the tonal range of a scene, but interpreting the effects of controlled lighting can be difficult or impossible. And they don't provide information in camera terms (i.e., f-stops)—that's the job of a light meter. (See Figures 34-36.)
Most camera manufacturers only give you an approximate representation of the subject histogram, as in Figure 35, because they realize the viewing window is too small to provide accurate detail. There are some exceptions to this, as in Figure 36. Cameras that offer more data are more expensive and/or are designed for the slower-paced studio environment where there is time to analyze and reanalyze both lighting and exposure.
In the end, the way you record your vision of a scene or subject is all about how you interpret and control the light falling on it, the combination of photographic techniques you use to capture it, and the care you use in exposing the image.
A handheld light meter enables you to see light more completely, and to make the types of decisions that will make you a better photographer.
Figure 35 Typical camera histogram.
Figure 36 High-end digital camera histogram (spot reading only available with proprietary software).
Challenge! Can you match each image to its histogram?
Trying to judge an image only based upon its histogram is like trying to tell what someone looks like based upon their fingerprint. Like a pilot’s instruments, a photographer’s light meter is the key to gaining valuable information about a scene’s tonal values and hues before it is captured. Give our matching quiz your best shot, then click here to find out which image goes with each histogram.