How do you discern whether you got what you were after? Let's face it, when you shoot a digital image, you immediately check the LCD display to assess your results. Do you keep the first image, or do you need to reshoot? How do you decide, especially if you're away from your computer or out in the field as I was while shooting in Tuscany? One thing is for sure: Don't trust the paltry little LCD screen on the back of your camera. It might show you composition and framing, but it lies and deceives when it comes to exposureit cannot show you anywhere near the level of detail necessary for critical decisions. The solution is to use histograms. A histogram is a graphic representation of how pixel information is distributed within the tonal range of an image. How much information is in the highlights? How about in the shadows? What's the darkest pixel? All these questions can be answered quickly using a histogram. When you're shooting in the field, you can access the histogram for the image you just shot to see whether you exposed properly for the highlights. If the highlights are clipped (too strong) or if there is no information in the highlight range, you know that the exposure is incorrect and you should probably reshoot. The next section explains how histograms are structured and how to read them so that you can make intelligent decisions about exposure and tonal range. Anatomy of a HistogramThe histogram is a common tool in assessing the tonal range of an image. It's in most camera preview systems, as well as in Photoshop and other image-editing software. Because the histogram is a standard measurement tool, you must know how to read a histogram correctly and understand how it relates to the image as well as your creative objectives. In a histogram, pixel values are represented along the horizontal baseline, increasing from left to right. Rising up from the baseline are peaks and valleys that graph the proportionate number of image pixels corresponding to the baseline value. A quick glance tells you whether you have any peaks in the highlights or shadows, signaling that pixel data is present in that area (see Figure ). Check to see whether image data is present across the full tonal spectrum, especially in the areas that interest you the most. A histogram and its corresponding tonal areas. For a scene that's evenly lit, a good histogram shows some pixel data at both ends of the spectrum, with the highest peaks (the most pixels) in the middle. Such a graph signifies an image with moderate highlights and shadows and the majority of information in the midrange. But what about that snow scene or that twilight image? The histograms associated with those images will look very different from the ideal ones. If there are no bright highlights in the scene, don't expect them to be in the histogram. This lecture might sound basic, if not remedial, but it reinforces the point that each histogram is unique, and there is no target histogram shape, just as there is no universal formula for perfect exposure. Histogram ShapesThere are as many histogram shapes as there are image types. Some emphasize the shadows, others the highlights. Don't fall into the trap of making all your image histograms look the same. What's more important is that you recognize the important areas in the image, and make sure that there is adequate pixel data in the corresponding area of the histogram. Figure shows a standard exposure, with bell-shaped histogram. Figure shows an image with the shadows dominant; notice that its histogram is skewed to the left. Figure shows an image with the highlights dominant; its histogram is skewed to the right. Figure shows an image dominated by a single flat color, with minimal highlights or shadows; its histogram is similarly flat (most of the pixels in the center of the histogram). Dried Flower II Yellow Flower Fresco Detail Table Cloth
|