Hi Kayron,
Thank you for writing this tutorial. I'd like to point out just a few things:
- Noise reduction should only be used for star detection in extreme cases, never on a regular basis. Basically, the purpose of this noise reduction parameter is to help with marginal data, where star detection has problems to distinguish among stars and noise. Applying star detection noise reduction when it isn't necessary may degrade registration accuracy.
- Defining previews is only necessary in difficult cases, where overlapping areas are very small. StarAlignment should have no problems at all to find the intersection between these mosaic frames automatically. This feature should only be used when necessary, not on a regular basis.
- Similarly, increasing the maximum number of distortion iterations is normally not necessary, except in difficult cases.
- I don't see the need to build a "rough mosaic", as you define it, to register mosaic frames over it. This double registration procedure can only degrade accuracy, especially on overlapped mosaic regions. Two-frame mosaics should always be generated in a single registration step. For mosaics of more than two frames, a much better option is to generate a synthetic reference image with the CatalogStarGenerator script, then use StarAlignment to build the mosaic on the synthetic image. Large wide-field mosaics where frames are highly distorted, very large mosaics consisting of many frames (6 or more), and mosaics with very small (or nonexistent) overlapping areas, are much better generated with the MosaicByCoordinates script, which is based on an astrometric solution.
- Finally, using the FITS format to save working images is a valid option, but I wouldn't recommend it in a tutorial. The FITS format has been deprecated in PixInsight 1.8.4, and will be supported exclusively for compatibility with existing data. I would recommend the use of XISF, which is and will be the native file format of the PixInsight platform.
