Unlimited AE and Premiere Pro templates, videos & more! Unlimited asset downloads! From $16.50/m
Advertisement
  1. Photo & Video
  2. Photography
Photography

Noise in Digital Photos: Where It Comes From and How to Fix It

by
Length:ShortLanguages:

Digital noise in photographs can actually appear a number of ways and be caused by different things. In this tutorial, we look at some of the major causes of image noise and how best to ‘treat’ them.

What is Digital Image Noise?

Noise is what we tend to refer to any digital ‘distortion’, most commonly grain.  All electronic devices create noise and just like audio noise, it creates unwanted extra bits of information caused by static. This is called the signal-to-noise ratio. Obviously we can’t hear image noise though.

Grain, ISO and Sensors

There are two types of this noise, chroma and luminance. Chroma noise causes a patchy, blotchy look to colour, and luminance noise makes a the image look grainy.

ISO is a measure of the gain, or amplification, your camera applies to your image to produce a useable image. Higher ISOs take the base exposure from your sensor and, like pushing up the gain of an electric guitar – the image becomes brighter but the rendition is less ‘clean’ and faithful to the original. This is a really great way of thinking about it because how it sounds is exactly how it looks, less clean. Take a regular picture on a sunny day at a very low ISO (100) and you get this:

picture at 100 iso
A picture taken on a sunny day at ISO 100

A very clean image with some slight, normal noise visible at 100%. At the other end of the spectrum, compare this with a picture on a regular sunny day at ISO 6400:

image taken at ISO 6400
An image taken on a sunny day at ISO 6400

You get a huge amount of noise at 100% and it’s also visible when zoomed right out. This would have been even worse on a crop-sensor camera as I'll explain shortly.

If you underexpose your image and then try and lighten it in post production, this is essentially like increasing the ISO but with greater effect. Say something is shot at ISO 800 but is still underexposed, it would look worse in post than if you’d shot correctly exposed at 1600. Ben Lucas has an excellent Quick Tip on how to nail your exposure

Other Causes of Grain

Sensor Characteristics

This is directly related to the ISO noise. Smaller sensors pick up less light so you have less flexibility with bumping up your ISO. Larger sensors pick up more light, which means they handle low light situations much better, so you can use a higher ISO and get less visible grain. However, a sensor with a lot of receptors packed onto it will also produce more noise. This is because the more receptors you have the smaller they have to be to all fit on the sensor, thus each collecting in less light.

Heat

Temperature can affect the noise produced, believe it or not. It means that we’re more likely to get grainier image with our DSLRs in summer than winter. This can also happen with long exposures as the camera is working away for long periods of time.

Static

When you have your shutter open for a while during a long exposure, this can cause a build-up of static which results in noise on your image. Many cameras have a noise reduction option specifically for long exposures but it does require the camera to fire twice, so you have to wait double the time.

Salt and Pepper Noise or Spike Noise

Sometimes when converting from analogue to digital, like scanning an image in, results in the appearance of white pixels in dark areas and vice versa.

salt and pepper noise
Salt and Pepper noise can occur when converting from analogue to digital

It’s hard to see here as it’s a good scan and by the very nature of an old photo, will already have some speckles anyway; but this is what it looks like.

Chromatic Aberration (CA)

Sometimes known as ‘fringing’, chromatic aberration is distortion caused by the lens this time, rather than the camera, but extra unwanted information nonetheless. As light passes through the glass and plastic elements in your lens the colours bend in different ways, and sometimes they can fail to transmit them all as they should and fail to bring them to the same focal plane. This manifests itself in lines of colour, usually along edges of objects, people or buildings because that’s where changes in colour naturally occur, where they go from light to dark.

chromatic aberration
An example of Axial Chromatic Aberration

There are two types of CA, Axial (or longitudinal) and Transverse (or lateral). Axial occurs when the colours don’t converge at the same point and results in fringing around edges throughout the whole image. It can be reduced greatly by using a smaller aperture so logically, faster prime lenses are actually more prone to this kind of CA.  

Transverse CA occurs when the colours do meet on the same focal plane but at different points, as they’ve come in at an angle. This fringing is only visible towards the corners of the picture and can’t be stopped by stopping down the lens. This is more common with wider lenses and those of a lower quality.

Noise Reduction and Removal

Noise can't be eliminated, it's always, at some level, part of photographs. However, in most cases it's possible to improve the appearance of noise so that it doesn't distract. We have a section of tutorials dedicated to noiseremoval and reduction including bespoke software packages and easy tweaks in Photoshop or Lightroom. Here are a few to get you started:

Conclusion

Digital image noise is bound to appear in some images. In most cases, it’s the cameras fault, not ours, hurrah! If you learn what causes certain types of noise though then you can take steps to reduce its appearance before you get to post-production.

  • High ISO causes grain but it’s still better to select a higher ISO in camera than correct an underexposed image in post
  • Your sensor size and characteristics will have an impact on how grainy something looks in low light. Smaller sensors can’t handle high ISOs.
  • Long exposures are often a cause of grain due to static and heat. Try using your in-camera Long Exposure Noise Removal tool.
  • When scanning in an image, do it at as high a quality as possible to give yourself the best chance of reducing salt and pepper grain
  • Check lens reviews before buying to see if they’re prone to CA; some cheaper ones are.

Finally, when noise does happen, don’t panic! Just take a look at the great range of tutorials we’ve got to help you reduce or remove noise (or even add more!) in post production.


Advertisement
Advertisement
Advertisement
Advertisement
Looking for something to help kick start your next project?
Envato Market has a range of items for sale to help get you started.