CMOS and sCMOS Camera Gain Settings: From Basics to Manual vs Auto |

When working with advanced imaging systems such as CMOS and sCMOS cameras, few settings are as influential as gain. Gain determines how signals from the sensor are amplified before being converted into digital values, directly shaping brightness, noise, and dynamic range. Yet, many users struggle with misconceptions about what gain really does, when to use manual versus automatic gain, and how to optimize it for their application.

This guide provides a clear, practical breakdown of what gain is, common misunderstandings, how it impacts image quality, and how to set it appropriately.

What Is Gain?

Camera system gain is the ratio of displayed gray levels to detected photo-electrons, measured in gray levels per electron. Sometimes, the inverse is provided — in electrons per gray level — but both describe the same relationship.

The exact gain value (or range of values) is set by camera designers through the analog-to-digital converters (ADCs), amplifiers, and capacitors in the readout architecture. This determines how many gray levels each photoelectron will be represented as, on top of the baseline offset. Gain also defines how much of the camera’s physical full well capacity is addressed within the available bit depth of different modes.

● Low gain: produces a darker but cleaner image with a wider dynamic range.
● High gain: brightens the image but introduces more noise and reduces dynamic range.

Figure 1: The effect of changing gain value

Depending upon the gain value, the exact same signal in photoelectrons can lead to significantly different gray level values. Without knowing the gain value, a gray level value is meaningless as a signal measurement.

Gain, therefore, determines the 'step size' of our signal intensity measurements – the precision with which photoelectron counts are digitally sampled. A simple analogy is audio: turning up the volume amplifies both the music and the background hiss. Similarly, in cameras, increasing gain amplifies both signal and noise.

Note: In consumer photography, gain is referred to as the “ISO setting.” This term originated from film photography, where ISO measured film sensitivity. Higher ISO numbers correspond to higher electronic gain in digital cameras.

Common Misunderstandings Around Gain

Although the term "gain" is familiar from audio or electronics, its use in imaging often leads to unhelpful assumptions. Misunderstandings can cause images to be misinterpreted or gain settings to be neglected.

1、"Gain is cheating."

The perception that increasing gain is somehow 'artificially boosting' signals is not true – increasing gain merely increases voltage measurement precision.

2、"1× gain means no gain."

The default gain setting of a camera, where multiple settings are available, still represents a chosen gain value in gray levels per electron. To say "this camera has no gain" is like saying "this person doesn’t have a height"! Gain is simply a measurable property of the camera’s operation.

3、"Higher gain makes signals brighter but noisier."

With the exception of EMCCD cameras, this is almost always false. Higher gain values, through multiplying signal and noise together, can simply reveal noise already present in images. Though, in fact, higher gain typically reduces read noise, and the highest gain setting a camera offers is usually the lowest noise.

How Gain Affects Image Quality

Gain settings influence three core aspects of image quality:

1、Brightness – Higher gain brightens images, particularly in low-light situations.
2、Noise – Amplifying weak signals also amplifies noise, including read noise and shot noise. At high gain, images may appear grainy.
3、Dynamic Range – Higher gain reduces the maximum range of signals the sensor can capture without saturating. This limits the ability to record both very bright and very faint details in the same image.

For CMOS cameras, gain may reduce effective dynamic range significantly at high settings. sCMOS cameras, thanks to their dual-gain architectures, often achieve lower noise while maintaining broader dynamic range, making them ideal for scientific imaging.

Setting Gain Appropriately

Figure 2: Setting gain appropriately

Top: Images captured with given gain settings.

Bottom: Image intensity histograms for top images.

Gain represents a key trade-off in scientific imaging: it determines how you balance sensitivity against dynamic range.

Increasing Gain:

● Reduces read noise, improving signal-to-noise ratio in low-light situations

● Improves quantization precision for weak signals (more gray levels per electron)

● Enhances contrast when imaging faint structures

Decreasing Gain:

Increases the available full well capacity, enabling capture of brighter signals without saturation

While not all cameras have changeable gain settings, many do to allow a balance between high dynamic range / full well capacity modes and high sensitivity modes.

Rule of thumb: Choose the highest gain setting (the most gray levels per electron) that you can, or the gain setting with the lowest read noise (if different), without coming close to saturating pixels in your signal of interest. If some pixels, due to the random variations of noise, reach the saturation value, then your gain may be too high if the data from these pixels is of importance.

Note: Be careful, however, as gain settings are sometimes tied into other camera modes, where changing modes changes not only gain but also bit depth, camera speed, or other operating modes of the camera.

Manual vs Auto Gain: Which Should You Use?

Aspect	Manual Gain	Auto Gain
Control	Full user control	Camera adjusts automatically
Consistency	High (reproducible across datasets)	Variable, may change frame to frame
Ease of use	Requires expertise	Simple and fast
Best for	Quantitative experiments, microscopy, astronomy	Live imaging, surveillance, dynamic lighting

Manual gain is preferred for scientific applications where reproducibility and quantitative accuracy are essential. Auto gain is convenient for real-time viewing or inspection tasks where lighting conditions fluctuate.

How to Discover Your Camera’s Gain Value

Knowing the actual value of camera gain in gray levels per electron is of great benefit in scientific imaging and is essential in some imaging applications. However, almost no camera software displays to the user the gain value of the camera in its current mode. There are a number of potential sources to discover this value:

1、Read the gain values for the different camera modes, as measured by the camera manufacturers, from certification documents that can come with scientific cameras.

2、Calculate approximate values from a camera specification sheet by dividing the full well capacity in each mode (if provided) by the maximum gray level value (given by the bit depth) available in that mode. Note however that specification sheet full well capacity values can occasionally be wildly overstated compared to real cameras, by as much as 40%. Every camera will have a slightly different full well capacity.

3、Measure the gain yourself with a mean variance test.

Gain Settings in Scientific Applications

Below is a table showing a suggested classification of gain values, and the corresponding full well capacity that could be addressed for 8 bit, 12 bit or 16 bit pixel values.

Table 1: Example gain values within typical range, in grays/e-

Example gain values and the corresponding inverse gain (in e-/gray), and the resulting maximum full well capacity that gain choice would access for a given bit depth (assuming no offset)

Conclusion

Gain is one of the most critical — and most misunderstood — parameters in CMOS and sCMOS imaging. It is not a magic tool for sensitivity, nor is higher always better. Instead, gain is a trade-off between brightness, noise, and dynamic range.

● Manual gain provides control and reproducibility, making it ideal for scientific and quantitative work.

● Auto gain offers convenience and adaptability, well-suited for live monitoring and variable conditions.

By understanding your camera’s gain values, avoiding common misconceptions, and applying best practices, you can optimize image quality while maintaining scientific rigor.