Image stacking is the single most powerful technique available to astrophotographers. By combining multiple exposures of the same target, stacking dramatically reduces noise, reveals faint detail that is invisible in any individual frame, and produces a final image of a quality that no single exposure — however long — can match. If you're new to astrophotography and wondering why your images look noisy and lacking in detail, stacking is almost certainly the answer. This guide explains what stacking is, why it works, and how to do it using the most popular free software tools. For a guide to capturing the best frames to stack, read: The Best Settings for Astrophotography with a Full Spectrum Camera.
What is Image Stacking and Why Does It Work?
When you take a single astrophotography exposure, the image contains two components: signal (the light from the stars and nebulae you're trying to capture) and noise (random variations in the image caused by the camera sensor, thermal effects, and other sources). In a single frame, noise can easily overwhelm faint signal, making it difficult or impossible to see fine detail in dim objects.
Stacking works by averaging multiple frames together. Because noise is random — it varies differently in each frame — it averages out when multiple frames are combined. Signal, on the other hand, is consistent across frames and adds together constructively. The result is that the signal-to-noise ratio improves with each additional frame added to the stack.
The improvement follows a square root relationship: doubling the number of frames improves the signal-to-noise ratio by a factor of approximately 1.4 (the square root of 2). To improve the SNR by a factor of 10, you need 100 frames. This is why serious astrophotographers capture many hours of data across multiple sessions — more frames always means a better result.
Types of Frames
A complete astrophotography dataset consists of several types of frames, each serving a specific purpose:
Light Frames
Light frames are your main science frames — the exposures of your target object. These are the frames that contain the signal you want to capture. Everything else in the workflow exists to improve the quality of your light frames.
Dark Frames
Dark frames are exposures taken with the lens cap on (no light reaching the sensor), at the same ISO, shutter speed, and temperature as your light frames. They capture the thermal noise pattern of your sensor — the fixed pattern of hot pixels and thermal current that the sensor produces regardless of the light falling on it. Subtracting dark frames from your light frames removes this thermal noise pattern, significantly cleaning up the image.
Dark frames should ideally be taken at the same temperature as your light frames, as thermal noise is temperature-dependent. Taking darks at the end of your imaging session, before bringing the camera indoors, gives the best temperature match.
Flat Frames
Flat frames are exposures of an evenly illuminated surface — typically a white t-shirt stretched over the lens and pointed at a bright, even sky, or a dedicated flat panel. They capture the vignetting pattern of your lens (the darkening towards the corners) and any dust spots on the sensor. Dividing your light frames by the flat frames corrects for these artefacts, producing an evenly illuminated image.
Flat frames must be taken with the same lens, aperture, and focus position as your light frames. They are typically taken at the end of the imaging session before moving the camera.
Bias Frames (or Offset Frames)
Bias frames are the shortest possible exposures taken with the lens cap on. They capture the base read noise and electronic offset of the sensor. For beginners, bias frames are the least critical calibration frame type — focus on getting good darks and flats first.
How Many Calibration Frames Do You Need?
- Dark frames: 20–30 frames is a good target. These can be reused across multiple sessions if the temperature and settings are the same.
- Flat frames: 20–30 frames per session. Flats must be retaken if you change the lens, aperture, or focus position.
- Bias frames: 50–100 frames. These can be reused across many sessions as they don't depend on temperature or exposure settings.
DeepSkyStacker — The Beginner's Choice (Windows)
DeepSkyStacker (DSS) is a free, open-source stacking application for Windows that has been the go-to tool for beginner and intermediate astrophotographers for many years. It's straightforward to use, handles RAW files from most cameras including full spectrum converted models, and produces good results with minimal configuration.
Basic DeepSkyStacker Workflow
- Open DeepSkyStacker and use the buttons on the left panel to load your frames: light frames, dark frames, flat frames, and bias frames.
- Check your frames — DSS will display a list of all loaded frames. Review them and uncheck any that are obviously poor quality (badly trailed, out of focus, or affected by cloud).
- Click Register checked pictures — DSS analyses each frame to identify stars, which it uses to align the frames. This step can take several minutes depending on the number of frames and your computer's speed.
- Click Stack checked pictures — DSS stacks the registered frames using your chosen stacking method. The default settings work well for most situations.
- Save the result — DSS saves the stacked image as a 16-bit TIFF file, which you then open in your image processing software (Lightroom, Photoshop, PixInsight, or Siril) for final processing.
Stacking Methods in DeepSkyStacker
For most beginners, Kappa-Sigma Clipping is the recommended choice — it effectively rejects outlier pixels (caused by cosmic rays, satellite trails, and hot pixels) while preserving the true signal. The default kappa value of 2.0 works well in most situations.
Siril — The More Powerful Free Alternative (Windows, Mac, Linux)
Siril is a free, open-source astrophotography processing application available for Windows, Mac, and Linux. It is more powerful and flexible than DeepSkyStacker, with a wider range of stacking options, built-in processing tools, and support for scripting to automate repetitive tasks. The learning curve is steeper than DSS, but Siril is capable of producing professional-quality results and is the tool of choice for many serious astrophotographers who don't want to invest in PixInsight.
Basic Siril Workflow
- Set up your working directory — Siril works with a specific folder structure. Create a main folder for your project, with subfolders named lights, darks, flats, and biases. Place your frames in the appropriate folders.
- Open Siril and set the working directory to your project folder using File → Set working directory.
- Use the Preprocessing Script — Siril includes built-in scripts that automate the calibration and stacking process. Go to Script → Execute script and choose the appropriate script for your setup (e.g., OSC_Preprocessing for one-shot colour cameras like your full spectrum converted mirrorless). The script will calibrate your light frames with your darks, flats, and biases, register them, and stack them automatically.
- Process the stacked image — Siril includes a range of processing tools including background extraction, colour calibration, histogram stretching, and noise reduction.
Other Stacking Software Worth Knowing
PixInsight
PixInsight is the professional standard for astrophotography processing and stacking. It is extraordinarily powerful, with a comprehensive suite of tools for every stage of the astrophotography workflow. The learning curve is steep and the cost is significant (€230 for a perpetual licence), but for serious astrophotographers it is unmatched.
Sequator (Windows, free)
Sequator is a simple, fast stacking tool specifically designed for Milky Way and wide-field astrophotography. It can stack star frames while keeping the foreground sharp — a technique called "ground and sky separation" — making it ideal for landscape astrophotography. For a full guide to Milky Way photography, read: How to Photograph the Milky Way with a Full Spectrum Camera.
Starry Landscape Stacker (Mac, paid)
The Mac equivalent of Sequator for Milky Way landscape stacking. Simple to use and produces excellent results for wide-field night sky photography.
Tips for Better Stacking Results
- Capture more frames — the single most effective way to improve your stacked result is to capture more light frames. Aim for at least 1–2 hours of total integration time as a minimum.
- Reject poor frames — frames affected by cloud, aircraft trails, or poor seeing should be excluded from the stack.
- Use calibration frames — darks and flats make a significant difference to the quality of your stacked result, particularly for removing hot pixels and vignetting.
- Shoot RAW — always. JPEG files discard too much data for stacking to be effective.
- Keep your camera cool — thermal noise increases with temperature. Imaging on cooler nights produces cleaner frames that stack better.
- Don't over-process — a well-stacked image with good calibration frames needs less aggressive processing to look good. Let the stacking do the heavy lifting.
Getting Started
If you're new to stacking, the recommended starting point is DeepSkyStacker (Windows) or Siril (all platforms). Both are free, well-documented, and supported by large communities of astrophotographers who have produced extensive tutorials and guides. YouTube is an excellent resource for step-by-step video tutorials for both applications.
Start simple — capture 20–30 light frames of a bright target like the Orion Nebula or the Pleiades, take some dark and flat frames, and run them through DeepSkyStacker or Siril. The improvement over a single frame will be immediately apparent and deeply satisfying.
Explore our range of full spectrum converted cameras to get the most out of your astrophotography stacking workflow — the improved hydrogen-alpha sensitivity of a full spectrum camera means more signal in every frame, and more signal means better stacks.