Closed EranOfek closed 1 day ago
It looks like the problem is in: imProc.psf.fwhm, while AstroPSF/fwhm results are fine. imProc.psf.fwhm uses AstroPSF/fwhm, but multiply it by the pixel scale. In images for wich the astrometry failed, the CD matrix diagonal contains 1, so the FWHM is of the order of 3600 X FWHM[pix]...
To fix it, now when we call imProc.psf.fwhm we give it also the 'Scale' parameter (1.25"/pix for LAST). [dev1 3facb4a67]
LAST pipeline: good images are rejected because FWHM is large (but it is not). Images are identified with FWHM of about 10,000 while the actual FWHM is 2-3 arcsec.