Fits header values after image solver script

fredvanner

Well-known member
Apr 17, 2019
86
4
Wells, Somerset, UK
... missing from my workflow above. Before saving as FITS, delete the PV1_1 and PV1_2 keywords. I suspect (but I am open to correction) that the value of PV1_2 is wrong. Most applications will assume the default values for the selected projection type if these keywords are missing.
These keywords are defined in Representations of celestial coordinates in FITS, where it is noted that:
"zenithal projections all have (phi_0; theta_0) = (0; 90)"​
Since the default TAN (gnomonic) projection is of zenithal type, I would expect PV1_2 = 90.
 

fredvanner

Well-known member
Apr 17, 2019
86
4
Wells, Somerset, UK
Consistent with the analysis above, I have confirmed that an imported FITS astrometric solution (e.g. from nova.astrometry.net) can be corrected to PI format by:
  • replacing CRPIX2 with NAXIS2 - CRPIX2 -1
  • changing the signs of CD1_2 and CD2_2
[The "-1" on CRPIX2 assumes indexing from 0..(NAXIS2-1); I confess my test can't check to that 1-pixel accuracy]
However, I can't think of a good reason for wanting to import a third party astrometric solution, unless all attempts at a solution in PI have failed. I strongly recommend re-solving any imported image with ImageSolver.