The Histogram and Picture Brightness
It’s helpful to review a photograph after taking it to ensure that your exposure is correct. In most circumstances, you can judge the exposure by looking at the preview picture on the LCD via the playback function. Nevertheless, there may be occasions when the lighting around you is bright and casts reflections on the small display, which can make judging a photograph’s exposure difficult. A more objective way to judge a photograph’s exposure is to refer to its histogram.
What is a histogram?
In photography, a histogram is a graphical representation of the number of pixels that occupy each tonal value on a grade from pure black to pure white. When looking at a histogram, the horizontal axis represents tonal gradation, and the vertical axis represents the relative quantity of pixels of a particular tonal value. Put differently, the height of the graph at any point on the horizontal axis represents the number of pixels of that tone.
In digital photography, an image histogram is always generated from a processed source image. Your camera generates histograms using the tonal variations in the JPEG format images, or the JPEG preview images attached to raw files. In most cameras and software suits, histograms aren’t generated from the unprocessed data in raw files.
Some mirrorless cameras can display a live histogram in the EVF or on the rear LCD during shooting mode, which can help you understand your scene and exposure settings before taking a photograph.
Interpreting the image histogram
In general, a histogram is useful only as a contextual guide. In the absence context, being either the source image or a good recollection of the composition, a histogram can only tell you how bright the resulting photograph is, but not how correctly it’s exposed. Although related, the concepts of brightness and exposure are not interchangeable. Whether a picture is dark, bright, or average depends both on the subject’s reflectance and the overall exposure.
For example, a correctly exposed photo of a black cat in a coal mine (low reflectivity) will produce a dark image with a correspondingly left-skewed histogram. A correctly exposed photo of a white dog on snow (high reflectivity) will create a bright image with a correspondingly right-skewed histogram. However, without the source images for reference, a histogram offers no clues about levels of reflectivity or exposures.
Darker or “low-key” photos produce histograms with graphs that are skewed towards the left. Brighter or “high-key” photos produce graphs that are skewed towards the right. Normals photos result in graphs that are spread about the middle because they feature subjects with a normal distribution of tonal values from bright, middle, and dark.
To reiterate, a histogram is not a reliable measure of overall exposure. Contrary to many guides for beginners, a correctly exposed photograph won’t necessarily produce a histogram with a loose bell-shaped curve extending outward from the middle.
Histograms and highlights
Histograms are useful for judging the exposure (and overexposure) of highlights. Highlights occupy the far right portion of a histogram’s tonal range. This is the portion of the histogram representing correctly-exposed bright subjects, such as white clouds, wedding dresses, and fluffy Samoyeds. When correctly exposed, light-toned subjects should look bright while retaining details and textures. When a light-toned subject is overexposed, it will appear completely white and featureless, and no amount of editing to reduce image brightness will recover the lost information. Overexposed highlights are frequently called “clipped” or “blown.”
Generally, the presence of some clipped highlights is no cause for worry, and they’re often unavoidable. For example, photos that incidentally include car headlights and bare lightbulbs, open flames, or the sun and its reflection on a polished surface will have some blown highlights, and that’s both acceptable and (mostly) unavoidable. Overexposed highlights are concerning when they occur on the main subject, such as a face in a portrait.
Contrast
You may determine the amount of contrast in a photograph by analyzing its histogram. Contrast describes the variation between the light and dark areas of a photograph. A low contrast image will result in a histogram with a large volume of pixels concentrated along a relatively narrow range of tones. A high contrast image will often produce a histogram with a broad distribution along the tonal range, or several narrow prominences set far apart.
Brightness and RGB histograms
Many cameras feature two types of histograms: brightness and RGB (red, green, and blue). Brightness histograms assign a brightness value to every pixel in a photograph and use that to produce a graph. The colour of individual pixels is ignored in favour of showing total scene brightness. However, these can be deceiving in the presence of vibrant, saturated subjects.
Digital cameras produce colour images by a clever combination of red, green, and blue light. Each digital image can be separated into its individual red, green, and blue constituents or “colour channels.” Each colour channel is a monochromatic representation of the brightness values for that colour.
An RGB histogram presents three separate histograms that plot the brightness values of the red, green, and blue colour values for every pixel in the photograph. RGB histograms are useful for troubleshooting overexposure in individual colour channels.