Reading the Histogram

One of the most misunderstood tools that most cameras, and every piece of post-processing software I’ve ever used, include is the histogram.  There are sometimes multiple histogram readouts available, which offer a ton of information about your image. However, the one I find the most helpful when it comes to judging exposure is the luminosity histogram, because it gives us helpful information about our images that can help us to make quick exposure adjustments on-the-fly.

A histogram is a statistical graph that gives a visual representation of the distribution of data. A typical luminosity histogram might look like this:

Histogram - Normal

Luminosity Histogram

From left to right, the histogram should be read as the amount of data contained in the image that contains shadow or dark tones (left-most side), midtones (middle), and highlights or bright tones (right-most side). Vertically, the histogram displays the intensity or amount of the particular tones. In the above histogram, there is a little bit of shadow data, very little highlight data, and an abundance of midtone data.

The image that corresponds to the above histogram looks like this:

Photo - Normal Histogram

Another way to look at the above histogram is that the left side represents pure black as the the data can be interpreted (numerically, in RGB values, 0 Red, 0 Blue, and 0 Green), while the right side represents pure white as the data can be interpreted (numerically 255 Red, 255 Blue, 255 Green). However, this information is not spread evenly across the histogram. Rather, it’s usually more along the lines of 25% shadow, 50% midtone, 25% highlight.

(Technical note: RGB values are numeric representations of the red, blue, and green values of a particular color, and are the de facto standard for measuring and representing these colors, although there are numerous standards that use different values and definitions, and thus represent the same values with different colors. The numbers have to do with the number of bits assigned to each value for a captured or sampled color. It’s a bit complex to explain the nature and value of bits to those not technically oriented, so it’s a bit outside the scope of the discussion, so just take my word for it that pure black is 0,0,0 and pure white is 255,255,255.)

Visually, that information might be represented as such:


So now that you know what this information represents, what does it actually do for a photographer? Well, since the left side is pure black (0,0,0), what happens to any shadow data that’s actually darker than pure black (off the chart on the left) is that it is effectively lost in the image, and is simply displayed as pure black, with the additional information discarded. This loss of data is known as clipping, because of the way the data is simply clipped off at a certain point, as if the extra information was cut away by a pair of scissors. You’ll typically see shadow clipping as a loss of detail in the shadow or dark portions of your image.

Likewise, since the right side is pure white (255,255,255), any information that is beyond those values is effectively lost and displayed as pure white. Clipped highlights might appear in your images as washed out blotches with no detail.

Clipped shadow data may also be referred to as dropped shadows, since the detail information below 0,0,0 is dropped out of the image. Clipped highlights are also referred to as blown highlights, since the detail in the area is blown out, replaced with a swatch of detail-less bright white.

Ideally, you’d want your histogram to look as the one displayed above: with all of the information clustered in between the left and right sides with a significant drop off of data before reaching the sides. A large amount of data on either end indicates a high likelihood that there is information present beyond the range of the graph, hence detail being lost due to clipping.

Not all clipping is necessarily bad for your image; it’s a matter of degree. Don’t let the presence of clipping on either side of the histogram dissuade you from keeping an image. While the histogram is an excellent guide for judging when your image is properly exposed, there are times when bunched up data on the far side of the histogram is to be expected or, sometimes, even desired.

When shooting in bright, sunny conditions, or shooting subjects with a lot of white present, it’s common for there to be a lot of data on the right side of the histogram. Think about this for a second… the histogram is simply a graphical representation for how much of an image resides in a particular tonal range. If you’re shooting a bright subject, you should expect there to be a lot of highlight information in the image.


Blown Highlights

The portion in the upper-left corner is what blown highlights will generally look like

Histogram - Light

Histogram showing blown highlights, as evidenced by the bunching of data at the right-most edge of the histogram


As you can see by the histogram, there are clearly clipped highlights in this image, as evidenced by the peak present against the right side of the histogram. In the image, this is mostly concentrated in the upper-left corner. That’s the sun creeping into the frame, as as you can see, it’s just pure white with no detail whatsoever. Simply put, the brightness of the sun overwhelmed the sensor, and the best it could estimate was pure white, with any detail above that completely obliterated.

Likewise, when shooting at night, you should expect a lot of shadow tones, and thusly, should expect to see a lot of data on the left side of the histogram.


Absolutely no detail in the shadow tones.

Histogram - Dark

Histogram showing dropped shadows, as evidenced by the bunching of data at the left-most edge of the histogram

As you can see, there’s a lot of black in that image of the moon, and the histogram shows a high concentration of data bunched up against the left side of the histogram. Clearly, there is a lot of data that’s not being shown in this image, due to it being clipped. However, it is not detail that would potentially add to the artistic effect of this image, so I was not too concerned with that data being lost. Granted, it’s not a terribly interesting image, but it lends itself well to demonstrating the point about clipped shadows.

Generally speaking, dropped shadow detail is not as critical in the final image as blown highlights. Dropped shadows tend to just be black portions, and can often be effectively used for effect, as in the case with silhouettes. Since our eyes are naturally drawn to the brightest portion of an image, blown highlights can be very distracting, and pull the viewers’ eyes away the artistic focal point that you’re trying to convey. In practice, it’s best to avoid blowing your highlights in an image, while I wouldn’t get overly concerned about dropping shadow detail, but it’s important to be aware of both.

The histogram is most useful when used on your camera’s LCD display while reviewing images you’ve just taken. The LCD is a VERY poor representation of the exposure of your image. Often, what looks great on your camera’s LCD might have major exposure issues when transferred to and viewed on your computer. NEVER use your camera’s LCD screen to judge proper exposure. The LCD is meant to judge proper composition, not exposure or color accuracy.

D700 Histogram Overlay

Histogram Overlay from Nikon D700

Whenever possible, use your camera’s luminosity histogram overlay to judge whether you’ve achieved a properly exposed image. Check your camera’s manual to find how to enable this feature.

Write a Comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>