S-distortion correction for 2D longslit spectra
siloubser wrote on Sep 25, 2006
Hi
How can I correct 2D images for S-distortion correction? I can trace the apertures with apall but then you have to extract 1D spectra for it to extract along the traced aperture. What I want to do is output a corrected 2D image (to feed into a code) - is this possible?
I also tried a sequence of identify, reidentify, fitcoords and then transform. But this doesn't work well if I only have one object on the slit (it will work if you have lots of stars etc).
How can I correct 2D images for S-distortion correction? I can trace the apertures with apall but then you have to extract 1D spectra for it to extract along the traced aperture. What I want to do is output a corrected 2D image (to feed into a code) - is this possible?
I also tried a sequence of identify, reidentify, fitcoords and then transform. But this doesn't work well if I only have one object on the slit (it will work if you have lots of stars etc).
Francisco Valdes wrote on Sep 25, 2006
Hi,
Indeed what you want to use is the LONGSLIT package and the sequence of IDENTIFY, REIDENTIFY, FITCOORDS, and TRANSFORM. Often times all you do have to trace the S-distortion is a single source. In that case the transformation will shift all the spatial cuts (columns/lines) to align the source you trace exactly along a column/line. If you want to determine a 2D spatial dependence of the distortion then you need to have multiple features to trace along the slit. This is often done with a special slit/slit mask. It is also sometimes done by moving a star along the slit in either a single exposure (closing the shutter between shifts)
or as multiple exposures. But basically you have to have taken data to map the distortion and if all you have is a single source there is not much more that you can do.
But again, you have correctly identified the path you have to use.
Yours,
Frank Valdes
Indeed what you want to use is the LONGSLIT package and the sequence of IDENTIFY, REIDENTIFY, FITCOORDS, and TRANSFORM. Often times all you do have to trace the S-distortion is a single source. In that case the transformation will shift all the spatial cuts (columns/lines) to align the source you trace exactly along a column/line. If you want to determine a 2D spatial dependence of the distortion then you need to have multiple features to trace along the slit. This is often done with a special slit/slit mask. It is also sometimes done by moving a star along the slit in either a single exposure (closing the shutter between shifts)
or as multiple exposures. But basically you have to have taken data to map the distortion and if all you have is a single source there is not much more that you can do.
But again, you have correctly identified the path you have to use.
Yours,
Frank Valdes
Last post on Sep 25, 2006