options for distortion polynomials

willemrood commented 2 years ago

Hi there,

package works well as intended although I have two suggestions/questions regarding fitting distortion polynomials. We use a wide FoV setup (24mm, 53x74 deg FoV) and we need at least a 6 order polynomial to fit the distortions for accurate wcs2pix conversions. The current solution outputs with a 2nd (?) order SIP correction and only limited options for fitting anything afterwards (stars list != target list).

Is it possible to export a list of targets and matched stars such that we can do our own distortion fitting, or
Is it possible to add a kwarg for setting the tune_up order polynomial? btw the link to the tweak2 method of astrometry.net seems to be dead in the documentation.

Kind regards

aMarcireau commented 2 years ago

Hi,

The Astrometry.net C API, which this package uses, does not provide access to the matched catalogue stars (at least to my knowledge). However, I should be able to add support for higher-order SIP corrections.

For reference, the tweak2 method is implemented here: https://github.com/dstndstn/astrometry.net/blob/47849f0443b890c4a875360f881d2e60d1cba630/solver/tweak2.c#L195

Kind regards, Alex

aMarcireau commented 2 years ago

Hi @willemrood,

I published a new version of Astrometry (https://pypi.org/project/astrometry/3.0.0/) with support for more solver options, including the SIP correction order (details here: https://github.com/neuromorphicsystems/astrometry#solutionparameters).

Can you let me know if that works as intended with your data?

neuromorphicsystems / astrometry

options for distortion polynomials #1