Lucky Imaging for the Moon: Capture and Stack Workflow with PIPP, AutoStakkert and RegiStax

Lucky imaging is the technique of capturing thousands of short-exposure video frames of a bright target (the Moon, planets) and stacking only the sharpest ones — the frames where atmospheric turbulence happened to be minimal. It is the standard method for producing high-resolution lunar and planetary images, and it works with surprisingly modest equipment.

This guide walks through the full workflow using three free tools: PIPP (pre-processing), AutoStakkert (stacking), and RegiStax (sharpening).

Why lucky imaging works

Earth's atmosphere is turbulent. At any given instant, turbulence distorts the image passing through your telescope. Over a long exposure, this blurs the image. But in a rapid-fire video stream, some frames catch moments of calm ("lucky" frames) where the atmospheric distortion is minimal.

By selecting and stacking only the best 10–30% of thousands of frames, you get an image that is:

• Much sharper than any single long exposure
• Far less noisy than any single short exposure (because you are stacking hundreds of frames)

Equipment needed

Step 1: Capture

Settings

• Resolution: Use the full sensor resolution (or crop to region of interest)
• Frame rate: As high as possible (30–150 fps depending on camera and brightness)
• Exposure: Short enough to freeze atmospheric motion (typically 1–15 ms for the Moon)
• Gain: Moderate — bright enough for a good histogram but not clipping highlights
• Format: AVI or SER (uncompressed or lossless)
• Duration: 30–90 seconds of video per capture (yields 1000–10000+ frames)

Tips

• Focus carefully. Use a Bahtinov mask or fine-tune manually using live-view magnification.
• Capture multiple overlapping video segments if you want to create a mosaic of the full disc.
• The best seeing is often in the hour after sunset or before sunrise, when thermal turbulence is lower.

Step 2: Pre-process with PIPP

PIPP (Planetary Imaging PreProcessor) is a free tool that prepares your video for stacking.

What PIPP does:

• Centres the target in every frame (important if tracking is imperfect)
• Optionally crops frames to a consistent size
• Removes bad frames (e.g., frames where the target drifted off-frame)
• Converts between video formats
• Splits a full-disc capture into overlapping panels for mosaic stacking

Basic PIPP workflow:

1. Load your AVI/SER file
2. Set the source type to "Solar/Lunar Full Disc" or "Close Up"
3. Enable "Object Detection" to centre the Moon in each frame
4. Set output format (AVI or SER — AutoStakkert reads both)
5. Process and save the stabilised output

Step 3: Stack with AutoStakkert

AutoStakkert is the standard free tool for planetary and lunar stacking. It selects the best frames and aligns them using an intelligent multi-point registration scheme.

Basic AutoStakkert workflow:

1. Open the PIPP-processed video
2. Analyse — AutoStakkert ranks frames by quality
3. Set alignment points (APs):
   • For the Moon, use a grid of alignment points (size 48–96 pixels)
   • Place them across the visible surface, especially on areas with detail (craters, ridges)
4. Set frame percentage:
   • Start with the best 10–25% of frames
   • More frames = smoother but slightly less sharp; fewer = sharper but noisier
5. Stack — AutoStakkert aligns the selected frames at each AP independently, then stitches the result into a single sharp image
6. Save as TIFF or FITS

Tips:

• Use smaller AP sizes for higher magnification close-ups; larger for full-disc
• If the result shows artefacts at AP boundaries, increase the AP size or use fewer points
• Stack at 1.5× or 2× drizzle if your sampling is undersampled (can recover additional resolution)

Step 4: Sharpen with RegiStax

RegiStax is primarily used for its wavelet sharpening module, which is highly effective on stacked planetary and lunar images.

Basic RegiStax workflow:

1. Open the stacked TIFF from AutoStakkert
2. Go to the Wavelets tab
3. Adjust the wavelet layers:
   • Layer 1 (finest detail): increase gently — this brings out the smallest structures
   • Layer 2–3 (medium detail): increase moderately — crater rims, ridges, and larger features
   • Layer 4–6 (coarse detail): usually leave minimal — these affect overall contrast
4. Use the Denoise slider on each layer to suppress noise amplification
5. Preview the result in real-time
6. Do Final to apply the sharpening
7. Save the result

Tips:

• Less is more. Over-sharpening creates ugly halos and artefacts.
• Apply sharpening to a stretched (non-linear) image — the wavelets work best on data that is already visually interpretable.
• Compare your result with the un-sharpened stack to make sure you are enhancing real detail, not noise.

Expected results

With good seeing and a 150–200 mm telescope, you can resolve:

• Lunar craters down to ~2–3 km across
• Rille systems and ridge structures on the mare surfaces
• Mountain shadows and central peaks in large craters
• Fine texture on the Maria

With larger apertures (250 mm+) and excellent seeing, sub-kilometre features become visible.

Common mistakes

FP Softlab context

FP Softlab's Moon3D tool lets you explore the same lunar surface features you will capture with lucky imaging. The gallery includes reference lunar imagery for comparison.

Further reading

• PIPP — Planetary Imaging PreProcessor
• AutoStakkert — Official Site
• RegiStax — Official Site
• FP Softlab — Moon3D