Loading...

For many photographers, particularly those early in the early stages of learning, it’s easy to get caught up in taking an exposure that looks good rather than an exposure that is good. Thankfully, that’s where the histogram comes in, and it really is a powerful tool that shouldn’t be overlooked.

In Issue 77 of Photography News, we took an in-depth look at mirrorless cameras. Among the many things they offer, many photographers find the live stream of information taken from the sensor to be one of the best – information like histogram data, for example.

Though some DSLRs do feature a Live Mode offering similar results, it’s typically reserved for more high-end models. One of the joys of virtually any mirrorless camera, however, is the ability to shoot using a live histogram, rather than taking a test shot, viewing it to check the data, then adjusting accordingly. In some instances, this can be the difference between nailing your exposure in one and missing the shot entirely.

But why would you need to use the histogram if your camera is equipped with Live View? Well, in many instances, the LCD or viewfinder alone can be deceptive. If there’s too much light on your screen or you’re in a very low-light environment, they may appear especially dark or bright. The same thing happens if your LCD brightness isn’t adjusted appropriately. Even in ideal conditions, your eyes will never be as accurate as cold, hard data.

How to use a camera’s histogram

If you’re going to use your camera’s histogram, first you’ll need a firm understanding of what exactly you’re looking at. Thankfully, it’s really very simple.

In the most basic terms, a histogram shows the amount of light present in a scene across the range, moving from black, to shadows, to midtones, to highlights and finally white, going from left to right on the graph. The height of the peaks shows the amount of data present in each area. For example, in a bright scene, you’ll see the majority of the data shift to the right, and the highest peaks will likely be seen there, too.

It’s well-advised to avoid ‘clipping’ where possible. This is where your exposure is balanced so that the data reaches one edge or the other, meaning you’ve hit complete black or complete white in the darkest or brightest areas of your frame. In these cases, data is lost and cannot be recovered in editing if needed.

Click the images to see a larger view

Looking at the white average histogram, we can see with the underexposed shot all the histogram data is left of centre with some clipping, while the opposite is true with the severely overexposed shot. In the balanced exposure’s histogram, the data is much more evenly dispersed and there’s virtually no clipping

It’s important to note that if you’re capturing Raw files, you’ll have additional leeway. The histogram that appears on your camera is very likely that of the JPEG, and what may be unrecoverable there may be well within the Raw file’s normal range.

Your camera may also feature an RGB histogram as well as the simpler luminance histogram outlined above. Where the luminance graph shows an average of all RGB channels combined, the RGB histogram shows data from the individual colour channels.

The data reads the same, but the three graphs will vary depending on how much of each colour is present in your scene. Some photographers prefer the greater detail, as it can prevent blown out colours in specific areas. An RGB histogram can show clipping on a certain channel, where a luminance histogram may not have due to the average being brought down by the other two channels.

Click the images to see a larger view

Comparing the white average histogram to the individual RGB channels in these shots, we can see that the correctly exposed shot’s average histogram looks far darker than the blue channel’s and even discounts some clipping. Similarly, the average histogram for the overexposed shot looks well-balanced, but the blue channel shows that this is again inaccurate

Histogram myths

The old adage states a correctly balanced exposure will show a smooth, mountain-like histogram. By meeting this standard, you can ensure you have a pleasing range of tones with good contrast, but realistically, this isn’t attainable in many cases.

There are countless things that go into making a photo great, and this is just one of them. You shouldn’t let a tool such as the histogram overrule your intention.

With certain scenes and photos, capturing irretrievably dark data might not matter, for example. If the intention is to capture a very dark image, there’s no chance you’d want that data back in any case, so the loss is no real loss at all.

Earlier, when we talked about using the histogram to avoid taking an exposure that looks good rather than an exposure that is good, what we didn’t mention is that ‘good’ is a relative term. It can only be determined by you in the moment of each and every shot.

Regardless of if you’re using it to get an accurately balanced exposure, or to ensure over- or underexposing exactly as you intend, the histogram can be extremely valuable. By using it as you shoot rather than leaving it as an afterthought, you may just get better results than ever.

Remember, you can always share your photos with us on social media by tagging @photonewspn.

MBP MPU banner
Pro Moviemaker MPU banner
Learn with us MPU banner

Sign up to Photography News Weekly

Get a weekly dose of the latest news, reviews and inspiration direct to your email inbox