Digital noise appears in photography as small specs throughout the image, especially in the shadows. Noise refers to unwanted information. It is similar to film grain in traditional photography and ranges from barely visible to annoying to a sign of low image quality.
It can become a glaring issue in extreme conditions. For example, low light situations such as night photography, long exposures, and shooting at high ISO settings produce more noise. The good news is that we have options on how to manage digital noise through camera settings, post-processing software, or a combination of both.
In this article, we’ll take a detailed look at a challenging issue in digital photography. We’ll define digital noise, the types of noise and their causes, and how to eliminate or minimize noise in your photos.
What is Noise in Digital Cameras?
In a digital image, noise appears as artifacts, not in the sense of a relic or antiquity, but as a random variation. It is a visual distortion that is not part of the original scene.
In the case of digital photography, it is produced by the image sensor or the electronics of the camera and tends to obscure image sharpness and clarity.
We compare the amount of photo noise to the amount of desired data as a signal-to-noise ratio. Photographers seek a high level of information or data compared to a low level of noise.
In good light, a quality camera can capture images with only a faint amount of noise. So it has a high signal-to-noise ratio. Less light, in most cases, leads to more noise.
An extreme example is radio astronomy, which has a very low signal-to-noise ratio. This form of photography operates in very low levels of actual light but employs sophisticated technology to reveal information and data from images that are mostly noise.
What are the Sources of Digital Noise?
There are a number of sources of digital noise in photography, including random noise, fixed pattern noise, and read noise.
- Random noise, such as photon shot noise or simply shot noise, is produced by natural fluctuations in light.
- Fixed pattern noise is due to variations in the sensitivity of the individual pixels in a camera’s sensor.
- Read noise, also known as banding noise, is electronic noise generated by the camera’s circuitry when it reads data from the sensor.
Any or all of these factors may contribute to noise in photography. However, they show up in your photographs as two types of noise: luminance noise and color noise.
The photo below of a blue sky with two detail windows illustrates luminance and color noise.
If you want to get deeper into the science of digital noise, including the difference between temporal noise and spatial noise, check this article in Image Engineering.
Luminance Noise vs. Color Noise
Luminance noise is a random noise associated with light. How much or how little light is captured by the image sensor determines the amount of noise in the image. It doesn’t influence the color tones of the image. However, it can affect the brightness and ruin an otherwise excellent photograph.
The manner in which a camera sensor responds to variances in light results in luminance noise. It leads to extraneous pixels becoming visible, creating a grainy effect in the image. It is most noticeable in the shadows and in low-light images.
Sometimes called chromatic noise or chrominance, color noise shows up as little fields of color in the single-tone areas of an image. This is also a random noise. It appears as dots, or pixels, of various colors and contrasts with other parts of the image.
Color noise looks spotted and blurred, with one color next to another one and then another scattered about in a random fashion. It is most visible in the darkest and lightest areas of the image.
How to Reduce Noise
While we may never eliminate noise from digital photos, we can take a number of steps to minimize it.
Start with an understanding of your camera’s exposure settings. Of the three points on the exposure triangle, the ISO setting impacts noise more than the shutter speed or aperture. Raising ISO introduces more noise into the image. But shooting at a slow shutter speed in low light settings also allows additional noise to accumulate on the sensor.
Shooting at a wider aperture allows more light into the camera and makes a lower ISO setting possible. At the same time, remember that the wider aperture limits the depth of field.
If you need a refresher, we have an excellent article on the exposure triangle.
In-Camera Noise Reduction
Many digital camera models offer built-in noise reduction (NR) for long exposure, high ISO, or both. This is in-camera software to identify and prevent noise and or minimize it.
But this comes at a price. Some detail is lost in noise reduction. As the camera works to reduce digital noise, the image takes on an increasingly artificial look. And if you choose the camera’s noise reduction setting, you are stuck with it. Photo processing software such as Photoshop and Lightroom allows for finding a better balance between noise and detail.
The ideal method to reduce digital noise is to minimize it with optimum exposure. Adjust the camera settings: shoot at a faster shutter speed and allow more light into the camera with a wider aperture. This will optimize the image’s exposure, avoid noise, and maintain a low ISO.
Shoot at the Camera’s Optimum ISO
A camera’s native ISO is the range of settings from lowest to highest. The ISO setting that delivers the best image quality is the one that doesn’t involve a voltage boost to the sensor. In other words, shoot at the ISO setting that doesn’t generate extra sensor heat and delivers the best signal-to-noise ratio.
Shooting in the golden hour, into light, captures shadows where noise is most likely, as in the photo below.
The owners manual will tell you the camera’s native ISO. Finding the optimum ISO setting, or range of settings that deliver the best results, may take a little work. But with little trial and error, you can discover the setting that keeps image sensor noise at a minimum while capturing the detail and tonal range of a better quality image.
How Does Sensor Size Affect Noise
The sensor size of a digital camera bears a direct impact on digital noise.
Among the primary distinctions between crop sensor and full frame cameras is how much noise they produce. The full frame sensor can capture more light; therefore, its signal-to-noise ratio is better than that of a crop sensor camera.
So the crop frame camera sensor generates more digital noise than its full frame brother.
Capture Images in RAW
If you shoot in JPG, that’s a choice you make. However, shooting in RAW captures more information in-camera. This gives you the greatest flexibility in post processing, the best approach to achieving less noise and more detail in your photos.
Blown-out highlights can be processed to recapture some of the detail. Shadows can be lightened, again, to restore lost detail.
And digital noise is more manageable, with greater post-processing options, if you shoot in RAW.
Managing Digital Noise After the Photo Shoot
It’s ideal to avoid noise; however, photo processing software helps fix noise or at least manage what we cannot avoid. The methods for dealing with noise vary with the type of noise.
Color noise involves de-saturating the image or reducing the color. That affects the color of the entire image, so a little of this goes a long way.
Luminance noise reduction results in a loss of sharpness throughout the entire image. So be careful when you reduce image noise levels so as not to lose too much sharpness or detail.
Noise Reduction in Lightroom
Removing or reducing digital camera image noise in Lightroom is fairly simple.
In the Develop module, scroll down the right side menu to Detail. Under Noise Reduction, you’ll find three sliders for Luminance and three for Color.
As you advance the Luminance noise reduction slider, you may want to adjust the Detail and Contrast sliders. This helps restore some of the detail and contrast lost when reducing the noise levels. Take care not to add too much noise reduction.
The Color slider is advanced to reduce color noise. Adjust the Detail and Smoothness sliders to soften the transitions in color tones. Again, be careful not to reduce color noise to the point of losing essential color in the image.
As you make adjustments, you can zoom in on the main window or view the zoomed-in image in the small window right under Detail.
The Detail module of Lightroom is shown at the right of the screenshot below.
Noise Reduction in Photoshop
Photoshop’s noise reduction process is a little more involved than Lightroom.
It offers the option of masking a layer where you are reducing noise. This limits color noise reduction to a specific area, so the entire image doesn’t get de-saturated. In addition, you can recover some apparent sharpness that was lost to luminance noise reduction adjustments.
Under Filter, select Camera Raw Filter. Click Detail to reveal sliders for Sharpening, Noise Reduction, and Color Noise Reduction. This gives you some adjustments that can improve image quality.
For a more advanced set of options, close Camera Raw Filter. Click Filter, then Noise, then Reduce Noise.
Start with Basic. Adjust the Strength slider to reduce digital noise. The Preserve Details slider works in conjunction with Strength to maintain sharpness.
As you adjust Reduce Color Noise, be mindful of de-saturation as the slider moves to the right. Sharpen Details impacts the total image, so don’t take it too far.
Click the Remove JPEG Artifact box. This minimizes the effects of JPEG compression, which include a blocky look, rough edges sometimes, called mosquito noise, and degraded color.
The screenshot below shows Photoshop’s Reduce Noise panel.
Photoshop’s Advanced Option
For additional sophistication in noise reduction, click the Advanced button in the Reduce Noise window. Then, you can select Overall, which is what we outlined above, or Per Channel.
Selecting Per Channel allows for adjustment to individual color channels. You can select Red, Green, or Blue.
For each color channel, you can adjust the Strength and Preserve Details sliders. This operation gives you even greater control of Photoshop’s noise reduction software.
Below is a screenshot of the advanced Reduce Noise panel in Photoshop.
Photoshop cannot remove noise completely; however, if you use it with finesse and a careful eye, you can achieve excellent results.
Other Software Options for Reducing Noise
The processes outlined above work well in most cases. However, if you want to take it a step further, a number of third party plug-ins work nicely with Photoshop.
– Denoise AI from Topaz Labs is a popular Photoshop plug-in and is highly rated by photographers.
– ON1 NoNoise AI also uses artificial intelligence to remove noise and enhance details.
– Dfine is available as part of the Nik Collection; you have to purchase the entire eight plug-in set.
– Neat Image is a low-cost solution for noise reduction in digital images. It comes as both a plug-in and a discrete program.
– Imagenomic offers a noise removal plug-in called Noiseware. Reviewers gave it ratings similar to DeNoise AI.
I have not used any of these plug-ins, so I’m not qualified to rate them. I’ve always been satisfied with the results I get from Photoshop and Lightroom.
Specialized Noise Reduction Techniques and Processes
Digital cameras generate banding noise, or read noise, through both the sensor and the camera’s electronics. And long exposures keep the sensor on for longer periods. As a result, additional noise caused by sensor heat appears in your photos.
Your camera model may have built-in long exposure noise reduction. A similar, and possibly better, result can be achieved through Dark Frame Subtraction.
This process requires the photographer to capture two images with the same camera, lens, and exposure settings, preferably at the same temperature. One frame is the photograph you want to be the end result, the light image. The other frame is a dark image taken with the lens cap on.
The two images are blended together in a subtractive mode. This uses noise from the dark image to remove the undesirable noise from the light image, removing fixed pattern noise and hot pixel noise resulting from low light and slower shutter speed.
Tom Migot produced a video on how to perform Dark Frame Subtraction in Photoshop.
Software for Night Sky Photo Noise
Several software programs specialize in removing digital noise in photography done under extreme conditions.
– DeepSkyStacker is a free program for photographers who work with the night sky and deep space images.
– Aries Productions developed an application for deep sky image processing called Astro Pixel Processor.
– PixInsight created a program for astrophotography and similar imaging professions.
Final Thoughts on Digital Noise
Noise in photography is a complex issue that needs to be addressed, especially as you enter the professional arena. Managing noise in your photos often involves a combination of capturing the lowest noise possible in-camera and making further reductions to noise levels in post processing.
Deciding how much noise is acceptable depends on the application. For example, when detail and tonal quality are a priority, such as in architectural photography or large-scale printing, a very low noise level is crucial.