What does the perfect histogram look like?
Like many new photographers, I remember asking this question when I purchased my first DSLR.
That complicated looking diagram flashing on the back of your LCD screen can be confusing and it’s easy to assume there is a correct way it should look. This is certainly a common misconception as there is no such thing as a “perfect histogram.”
What is a histogram?
A histogram is a graph that represents the tones in your image. It reads from left to right with the darker pixels represented on the left, and lighter pixels represented on the right.
The progression from left to right reads: shadows > midtones > highlights. Essentially, a histogram is a tool that helps you determine if your image is properly exposed. It should be a representation of the tones in a given scene.
One important thing to note: If you see a section of the histogram that is climbing the wall on the left, it may mean you’ve underexposed the image or that a part of the image is underexposed. If you see a section of the histogram that is climbing the wall to the right, it may mean you’ve overexposed the image or that some areas of the image are blown.
Let’s look at a few photo examples…
Ideally, an image will have some pixel representation spanning across the entire histogram. This image is fairly “typical” in that much of the pixels are located in the mid-tones.
The part of the graph to the left is representing the darker flowers at the top left and the darker shadows in her sunglasses. This histogram makes sense to me based on the image it represents.
Now without even looking at the histogram we can tell that this image should have more pixels to the right due to all the bright snow. When I took this image I looked at the back of my screen to ensure the histogram wasn’t climbing the right wall. I wanted the snow to be bright but not overexposed.
This histogram looks entirely different from the previous but both are still “correct.” If you look closely at the histogram you will see that there are pixels spanning the entire graph while most of them are at the left. This makes sense when you look at the image. The majority of this image is darker with a smaller representation of lighter tones in the dew drops.
In this example, you can see that some of the pixels are climbing the right wall. This indicates that some part of the image is overexposed.
If you use Lightroom you have the option to turn on the “Clipping Warning” to see which part of your image is being affected. If you look to the top of the histogram in Lightroom you’ll see small arrows. If you click on those it will highlight in red which part of the image is impacted.
In this case, it’s only the windows that are blown. Since it’s not an important part of the image it’s not a concern to me. The rest of the scene is properly exposed.
In this example, a part of the histogram is climbing the wall to the left. When I turn on the clipping warning in Lightroom it shows in blue that some of the shadowed area to the right of her face here is clipped and underexposed. Given it’s not an important part of the image and the rest of the scene is properly exposed it’s not a concern.
For comparison, here’s a SOOC (straight out of camera) image that was overexposed and it does matter. This image was one of the first I took with a brand new camera and hadn’t quite figured it out yet.
As you can see, the histogram is climbing the wall to the right and when I turn on the warnings in Lightroom you can see it’s not only the window that is overexposed, but also much of her skin as well. This is where referencing the histogram on the back of your camera is really important; once I saw this I was able to adjust my settings and properly expose the image.
Here’s another SOOC image. I love using the histogram to determine whether an image would be a good candidate for a black and white conversion.
An image is generally a good choice for a black and white when it has a large peak somewhere in the histogram. When I saw her standing in this section of brighter light I knew I would convert this image. The light highlights her body perfectly and the rest of the image falls into the shadows.
This image we discussed earlier would also be a great choice for a black and white. The histogram has a large peak to the left with enough pixels represented throughout the rest of the graph. While ultimately I do prefer this one in color, it does make a strong conversion.
Remember, just as every image is unique, every histogram is as well. There is no “perfect” histogram.