Calibration, plate solving and quality checks for member-submitted images

1. Purpose

Members of ASRU image the sky with a wide range of equipment, from camera lenses on tripods to cooled astronomy cameras on small telescopes. Before their images can be used for anything else, they need to be calibrated in a consistent way, placed accurately on the sky, and checked for quality. This note describes how the pipeline does that.

2. Inputs

A submission is one observing session from one setup. It contains:

• light frames in FITS or a common camera raw format,
• bias, dark and flat frames taken with the same setup, and
• a short form with the camera, the focal length and, if known, the pixel scale.

Raw camera files are converted to FITS. Colour images from one-shot-colour sensors are kept as raw Bayer data until after calibration.

3. Processing steps

1. Checks. File formats, exposure times, temperatures and header keywords are checked. Frames with missing or inconsistent metadata are flagged rather than rejected.
2. Calibration. Master bias, dark and flat frames are built by median combination and applied using ccdproc. Dark frames are scaled by exposure time only when no matching exposure is available.
3. Plate solving. Each calibrated frame is solved with a local installation of the astrometry.net solver (Lang et al. 2010), using index files matched to the frame’s field of view. The pixel scale from the form, if provided, is used as a hint.
4. Source extraction. Sources are extracted with SEP (Barbary 2016), a Python implementation of the core of Source Extractor (Bertin & Arnouts 1996).
5. Cross-match and photometric zero point. Sources are matched to Gaia DR3 (Gaia Collaboration 2023) and a zero point is fitted in the Gaia G band.
6. Quality metrics. For each frame we record the median full width at half maximum, star elongation, sky background, number of detected sources and the fraction matched to Gaia.
7. Report. The submitter receives a summary for each frame and for the session as a whole.

4. Test results

We tested the pipeline on 412 frames from nine member setups, taken between November 2025 and January 2026.

Outcome	Frames	Share
Solved on first attempt	349	84.7%
Solved after a pixel-scale hint was added	22	5.3%
Not solved	41	10.0%

Of the 41 frames that were not solved, 19 had too few stars (mostly cloud), 13 had wrong or missing focal-length information, and 9 were badly trailed because of guiding problems.

For solved frames, the median residual between our positions and Gaia positions was 0.6 pixels.

5. Limitations

• Flat frames taken at a different time or with a different focus position are accepted but can leave visible patterns. The report flags large gradients but does not correct them.
• Very wide-field images (focal length below about 20 mm) are often not solved because of lens distortion. A separate path for these frames is planned.
• At the time of writing, satellite trails are not masked. This will change when trailmask is ready.

6. Availability

The pipeline code is released under the MIT license. Only members can submit images at present.

Acknowledgments

This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement. This work made use of Astropy, a community-developed core Python package and an ecosystem of tools and resources for astronomy (Astropy Collaboration 2022).

References

• Astropy Collaboration et al. 2022, ApJ, 935, 167
• Barbary, K. 2016, Journal of Open Source Software, 1(6), 58
• Bertin, E. & Arnouts, S. 1996, A&AS, 117, 393
• Gaia Collaboration, Vallenari, A., et al. 2023, A&A, 674, A1
• Lang, D., Hogg, D. W., Mierle, K., Blanton, M. & Roweis, S. 2010, AJ, 139, 1782