Exposure time, stacking and image noise

Signal, Noise and Signal-to-Noise Ratio

Astrophotography is all about collecting as much useful light from a celestial object as possible. This light forms the signal. At the same time, every exposure contains unwanted fluctuations and disturbances known as noise. The fainter an object is, the more important the relationship between the actual signal and this noise becomes.

The signal-to-noise ratio, commonly abbreviated as SNR, describes how clearly the desired image signal stands out from random fluctuations. A higher signal-to-noise ratio makes faint structures easier to reveal and allows an image to be processed more extensively without unwanted noise becoming overwhelming.

Where Does Image Noise Come From?

Astronomical images contain several sources of noise. The most important include photon noise, read noise, dark current noise and noise from the sky background.

Photon noise results from statistical fluctuations in the number of incoming photons. Even a perfect camera would be affected by this type of noise. It cannot be eliminated completely, but collecting more light reduces its relative impact on the signal.

Read noise is introduced when the camera sensor is read out electronically. Dark current noise is associated with thermal activity within the sensor. Light pollution, moonlight and other background sources also contribute additional noise.

For faint nebulae and galaxies, the signal from the object may be only slightly stronger than the sky background. This is why these targets benefit particularly from long total integration times and the darkest possible observing location.

Why Longer Total Integration Times Help

A single short exposure often contains only a small amount of light from a faint celestial object. By collecting many exposures and combining them afterwards, the total amount of available signal increases.

This process is known as stacking. The individual images are first aligned and then combined mathematically. Random noise is reduced relative to the useful signal, while recurring structures belonging to the celestial object become more clearly visible.

Under comparable conditions, the signal-to-noise ratio improves approximately with the square root of the number of combined exposures.

For example:

To double the signal-to-noise ratio, you therefore need approximately four times the total integration time under comparable conditions.

Individual Exposure Time and Total Integration Time

When planning an astrophotography session, it is important to distinguish between individual exposure time and total integration time. Individual exposure time describes how long a single image is exposed. Total integration time is the combined exposure time of all usable individual images. For example, collecting 300 exposures of 20 seconds each results in a total integration time of 100 minutes. Many short exposures can produce excellent results under suitable conditions. However, very short exposures may be more strongly affected by repeated read noise. Excessively long individual exposures, on the other hand, increase the risk of saturated stars and unusable frames caused by tracking errors or other disturbances.

The key is therefore not simply to use the longest possible individual exposures, but to combine an appropriate exposure length with a sufficiently long total integration time.

Filters and Their Effects in Astrophotography

Filters can be used very selectively in astrophotography to emphasise particular wavelengths or suppress unwanted light. The most suitable filter always depends on the target, the observing location and the current conditions. A filter does not automatically improve every image – what matters is whether it is appropriate for the object being photographed.

Broadband filters transmit a large portion of the visible spectrum. They are particularly suitable for objects whose light is distributed across many wavelengths, such as galaxies, star clusters and reflection nebulae. For these targets, preserving as much of their natural light as possible is important.

Narrowband filters, by contrast, transmit only very narrow wavelength ranges. They are mainly used for emission nebulae, which radiate strongly at specific spectral lines. Typical examples include H-alpha, O III and S II. These filters can make it easier to distinguish faint nebular structures from the sky background.

Dual-band filters combine two important emission regions, usually H-alpha and O III. They are particularly practical because many emission nebulae can be photographed effectively without using multiple separate filters. This can provide a significant advantage under light-polluted skies.

UV/IR-cut filters block ultraviolet and infrared light outside the visible spectrum. If a camera system is sensitive to these wavelengths, such a filter can help produce sharper stars and reduce unwanted colour fringes.