PCPhoto 
Learning Center The Magic Of Histograms
Learning Center The Magic Of Histograms
The Magic Of HistogramsLearn to read your camera's histograms and get the best exposures possible for your subject |
|
|
| By Rob Sheppard, Photography by Rob Sheppard | |
|
Page 3 of 4 That’s not necessarily wrong, however. Some scenes simply have too great a contrast range for your sensor, which will result in clipping at either the left or right (or both) because it’s impossible for your camera to do anything else. You may have to decide what’s most important for your scene—highlights or shadows—and be sure the histogram isn’t clipped at either end. Bright specular highlights from reflections of glaring lights probably should be pure white, for example. Or the darkness of a deep shadow might look just fine as black. Underexposure will show up with little or no data on the right side of the chart, while the rest of the data is skewed to the left. Often, the hill at the left will be cut off sharply at the far left edge. This can cause two major problems. The first is obvious—less detail in the shadows. The second appears when you try to adjust the image—noise. Underexposed, dark areas often will pick up annoying and distracting noise. The correction is fairly easy: add exposure. If you’re on automatic mode, use the + (plus) part of exposure compensation to bump up exposure. Overexposure demonstrates the opposite. You’ll find little or no data on the left side of the chart compared with the right, and the right side often will be cut off sharply. That cutoff is a real concern, since those areas of exposure beyond that level will be washed out, appearing blank in the photograph. The correction: subtract exposure. If you’re on automatic mode, use the – (minus) part of exposure compensation to lower exposure. A low-contrast scene shows something entirely different. Typically, it will show a histogram with an entire hill of data well within the left and right borders. No tonal values will be seen near either the left or right side. This can be a challenging histogram because often it will need to be stretched in your image-processing program. That stretching can cause problems with tonalities banding because not enough data may be present to support all the tones.
 |
