Noise Explained in Photography: The Screaming Camera

0
6

Noise is a topic in photography that seems made to cause confusion. Case in point: Several years before I bought my first DSLR, I had a point-and-shoot that I really wanted to learn how to use – but I was clueless about photography. When I read online that a high ISO setting “adds more noise” to a photo, naturally, I started thinking that a camera actually grows louder at those settings. I tested this theory by taking two photos at different ISO values, and – I could have sworn it! – the camera’s shutter was significantly louder at the higher ISO. For an embarrassingly long time afterwards, I went around thinking that high ISO values were fine to use, except in museums or cathedrals where silence was required. I doubt that many other people have been so hopelessly misguided about noise, but this article is worth writing regardless. Indeed, there are several aspects of noise that even advanced photographers often misunderstand. So, what is noise in photography, really, and what can you do to minimize it?

What Is Photographic Noise?

At some level, we’re all quite familiar with the concept of noise – if not in photography, then in other fields, such as music and audio recording.

You’ve surely noticed that, even in a quiet room, there is a background “hiss” in videos or audio that you record. That hiss isn’t something we hear normally, but it shows up in audio recordings (especially with a lower-quality microphone). Somewhere along the way, imperfections crept into your sound.

Aliexpress INT

The same is true in photography. In fact, even if you take a photo with your lens cap on, the resulting picture won’t be totally black. It might be close, but there will always be tiny imperfections: random, bright, and discolored pixels.

In this case, you can see the random pixels very easily just by brightening the image in Lightroom or Photoshop. If you’ve never done this before, it’s reasonable think that it would simply scale a photo smoothly from black to gray to white without an issue – but that’s not the case. Instead, in practice, the photo will become uglier and uglier, with huge areas of discoloration and strange-looking pixels. Welcome to the magical world of noise.

Noise With Lens Cap On
These random imperfections are called noise.

Disappointingly, noise is not the actual loudness of your camera; it matters much more than that. Noise shows up as a grainy veil in an image, obscuring details and making the picture appear significantly worse. In some cases, photos can be so noisy that they are essentially unusable. This article shows you what to do about it.

High levels of noise
This photo, taken at ISO 12,800, has a tremendous amount of visible noise. This is way too much for any reasonable uses.
What Causes Noise in Your Photos?

Technically, some amount of noise will always be in every photo. There’s nothing you can do to prevent this; it’s a physical property of photography.

There are two broad types of noise in your photographs: photon noise and electronic noise. Although they come from different sources, photon noise and electronic noise are typically hard to distinguish from one another when you look at the final photo, since they generally lead to the same result: pixels that are randomly too bright, too dark, or discolored.

Photon noise is randomness caused because the real world is random. Photons of light are emitted or reflected off of everything that we can see, but it doesn’t usually happen in a fixed pattern. For example, a dim lightbulb may emit an average of 1000 photons per second, but each individual second will be a bit different — 986 photons, 1028 photons, 966 photons, 981 photons, 1039 photons, and so on. If you’re taking a one-second long picture of this lightbulb, you won’t get exactly the same result each time.

Electronic noise is randomness caused by your camera sensor and internal electronics. Sometimes, this will have a clearly visible pattern, although the degree to which this is true depends upon the camera.

Both photon noise and electronic noise are important. Photon noise typically has a greater effect on your photos, but electronic noise is the reason why a lens-cap photo isn’t completely black. Each makes a difference.

Landscape at high ISO with some noise
NIKON D800E + 20mm f/1.8 @ 20mm, ISO 3200, 1/20, f/4.0

How to Minimize Noise

You can think of noise as, essentially, a “backdrop” for every picture you take. It will always be there, no matter what you’re photographing. Your goal, then, is to have the actual data (i.e., the real scene you’re trying to photograph) overpower this background. The best way to do that is to capture more light.

Consider a situation where you don’t capture enough light in the field, and the noise in an image overpowers the signal – the actual information. First of all, your photo will be extremely dark. You didn’t capture much light from the scene. But beyond that, when you attempt to brighten the photo on your computer, you’ll make both the signal and the large proportion of noise more visible, resulting in a photo that looks hugely grainy and discolored!

If you’ve ever heard the term signal-to-noise ratio, this is what it’s referring to. A photo with “more noise” isn’t always a bad thing for image quality – because the signal might have increased as well, perhaps by a proportionally greater amount, making the noise less visible overall. What matters here is simply the ratio.

So, how do you get the best image quality in your photos? It’s all about capturing more actual signal so that you can overpower the backdrop of noise that will always be present. You can do this by using a longer shutter speed, setting a wider f-stop, or photographing a more luminous (brighter) scene. In other words, by capturing a greater “luminous exposure.”

That’s how you reduce the appearance of noise in an image. Anyone who tells you to use a lower ISO to reduce noise is oversimplifying things. If you just lower your ISO without changing any other settings to capture more light, you’ll simply get a darker photo – a photo which you need to brighten in post-processing, revealing all the noise you tried to hide (and, in fact, typically more than if you had just used a higher ISO).

Base ISO and More Exposure
NIKON D7000 + 105mm f/2.8 @ 105mm, ISO 100, 1/40, f/3.2
Here, I was able to capture a lot of light with my aperture and shutter speed, ensuring maximum image quality.

How ISO Affects Noise

Your ISO is the only camera setting other than aperture and shutter speed that brightens a photo. Commonly, raising your ISO (to get a brighter photo) is said to increase noise. Is that true?

Start with the basics. ISO has no effect whatsoever on photon noise. It physically can’t. As we covered a moment ago, photon noise is entirely about the randomness of light emitted and reflected from the scene itself – something that couldn’t possibly depend upon your camera settings.

So, ISO only affects electronic noise. And the way it affects it will be surprising, at first, but it makes sense after some thought: For typical cameras at normal settings, raising your ISO will lower the amount of electronic noise. Yes – exactly the opposite of what you’ve probably been told.

Before you quit photography in exasperation, remember: What matters for image quality isn’t the actual amount of noise. It’s the signal-to-noise ratio.

Using a higher ISO will reduce the “amount” of noise. But when you’re shooting at a high ISO, it’s because you had no choice and couldn’t brighten the photo any other way – i.e., by capturing more actual light. This means that your signal-to-noise ratio won’t be very good. In other words, you didn’t capture enough data to overpower the curtain of noise, even if that curtain is slightly less strong.

I’ll emphasize here that it’s a good thing for your camera to reduce electronic noise at higher ISOs. That’s why we raise ISO in camera rather than shooting at base ISO and brightening everything in post-processing – you get an image quality boost that way, since electronic noise is lower. But signal-to-noise ratio is what really matters for image quality, which is why photographers don’t go around shooting everything at ISO 12,800 all the time. By far the best way to reduce the appearance of noise in an image is to drown it out with light.

(Indeed, along the same seemingly-insane-but-actually-makes-sense lines, taking photos of a more luminous scene will increase the “amount” of photon noise. But it increases the signal far more, improving your signal-to-noise ratio, and thus image quality. I’ll happily go into more detail about this in the comments if anyone has questions, including the background formulas.)

Milky Way with some noise
NIKON D800E + 14-24mm f/2.8 @ 14mm, ISO 3200, 25 seconds, f/2.8
There’s a reason I used ISO 3200 here rather than brightening a low-ISO photo in Lightroom or Photoshop: The image quality is better, thanks to the lower level of electronic noise.

Does Noise Reduction Software Work?

Finally, some people certainly will wonder about “noise reduction” settings in their post-processing software. Do these actually reduce noise, or is there a catch?

Unsurprisingly, there’s a catch. Using noise reduction algorithms will reduce the apparent noise in your photo, but it also harms legitimate details and makes them less sharp. If you use too much noise reduction, you’ll end up with photos that look like plastic. That’s far worse than some simple grain.

Noise reduction is still a useful tool. If the noise in an image is especially obvious, you’ll want to use post-processing to reduce some of it. A useful technique here is to apply selective noise reduction to large areas without much detail, like out-of-focus backgrounds, while reducing noise to a smaller degree on the image as a whole. (We’ve written about some of these noise reduction techniques before as well.)

The bottom line: Don’t avoid noise reduction entirely, but be cautious when you use it.

Some noise at high ISO
NIKON D7000 + 105mm f/2.8 @ 105mm, ISO 3200, 1/100, f/2.8
I used noise reduction here, including local adjustments to improve the quality of the background. This is especially noticeable at larger print sizes.
Facebook Fans