Noise is a random and unwanted problem caused by the sensor (determined by your ISO settings) in digital cameras. Most often, noise appears as randomly spaced grain or dots distributed throughout your image. Noise can also appear as unevenness in color, random pixels of color or as a variation in brightness within your image.
It’s most noticeable in plain, solid areas of your digital image, is also found in areas that are highly saturated and in blacks and shadows. The lack of light in these images means the sensor is working harder to distinguish detail in these areas. With a high enough ISO your sensor starts to get a little creative with where it notices color. “Sure the shadow is black, but that part is more red-black, and this pixel is more blue-black.” your sensor says. While back in the office, you beat your head on your desk over your sensor’s immaturity and lack of discernment.
Consequences of Noise
Noise is annoying. It distracts from your composition and causes your eye to jump around the image attracted to all these little spots of color. The presence of noise causes your image to have reduced image detail and clarity. It removes the focus from the subject of your image by creating an inconsistency in darker areas of the image, leaving these areas looking grainy and often with random color specks. This can confuse your image and your message.
How to Spot Noise
If you shoot with a High ISO and you’ve ever noticed a colored fuzz over your image, you’ve seen noise. The true effects of noise may be hard to notice at first. But zoom in on your digital image – 100% or more – you’ll see it’s there lurking the the darker areas or your image, like a pixel army ready to destroy your edit. In your raw image you may not notice it very much, but as you edit your image, the noise gets edited as well. Brighten those darks and you’ll see it. Sharpen your image just a little and you’ll see it as well. Effectively any edit will accentuate the noise in your image. There are generally two types of noise that you will see in your digital images: Luminance Noise – the contrasty dots or “digital grain” in an image, and Chrominance Noise – the pastel-colored, speckled noise often seen in mid-tone or shadow areas.
What Causes Digital Image Noise?
Noise can be a result of a number of things including:
- A very sensitive sensor.
- Low light combined with a high ISO.
- Your sensor size.
- Underexposure (Improperly exposing your image to be too dark.)
What is ISO?
The ISO setting defines how sensitive your camera sensor is to the amount of light present in your scene. A higher ISO setting will cause the sensor to be more sensitive allowing you to take pictures in low-light or with faster shutter speeds. A lower ISO makes your sensor less sensitive to light, causing it to be a bit more astute about it’s impressions of color in your image.
It’s always best to use the lowest ISO setting possible for proper exposure. However, as in all things, exposing an image properly is a balance between several factors: your creative vision, the type of image your trying to capture and the environment it’s captured in. Having a higher ISO enables you to:
- Reduce image blur when shooting hand held by using a higher shutter speed.
- Faster shutter speeds to freeze motion. This works great for sports photography or any photos where your subject is moving quickly.
- Get better performance in low light. Having a high ISO is generally needed for any kind of low light photography.
- Use a higher aperture setting for a deeper depth of field. This works great for landscape photography to capture more of the landscape in focus.
Noise Removal Tools
The good news is that there is no need to give yourself a concussion back in the editing room. Topaz DeNoise uses the most advanced noise reduction algorithms to reduce high ISO image noise while preserving and even extracting important image detail. Other noise reduction tools come with the unwanted side effect of detail loss, negatively affecting the overall quality of your image by blurring pixels together. With Topaz DeNoise, a custom noise profile for each image is created, allowing it to intuitively detect the difference between image noise and the details you want to preserve.