DADS will run 'elixir -D mode detrend big.list'
big.list contains lines of (/fullpath1/input.fits) (/fullpath2/output) (mode)
mode = split or mef
the detrend elixir finds the appropriate detrend data, applies it, and writes the resulting image as a SPLIT or MEF format file. If it is a MEF image, the file is given the name /fullpath2/output.fits. If it is a SPLIT image, the files are given names in the form /fullpath2/ouput/outputNN.fits, with NN = 00 - 11. The input images are always OBJECT files (science sky images), and have names at CFHT of the form 555555o.fits. To distinguish the image type as a processed OBJECT image, and to maintain consistency with the CFHT image naming scheme, we will convert the 'o' to a capital 'O' in the output files: 555555O.fits or 555555O/555555O00.fits.
In addition to the detrended images, the detrend elixir also produces a set of reduced data products. For the moment, these consist of jpeg versions of the images scaled to have a good color range. These can be used by DADS to produce html reports for each output image. These jpeg images are created in two forms: one binned 10x10, and one binned 50x50. These get names of the form /fullpath2/output.b10.jpg and /fullpath2/output.b50.jpg respectively.
Finally, the detrend elixir also writes the astrometric information determined by Elixir to the headers of the images which are created.
The archive process should halt and send out a warning message if any of the requested images fail this process. The likely causes of failure include:
These failures should be fairly uncommon, and can be handled by hand when they arise. In a short time, we will implement the detrending of subrastered images, and soon after than we will implement the detrending of binned images. In the latter case, the question at hand is if we can use re-binned detrend images on a on-chip binned target image. There may be small differences in reponse which may make this impossible. In the near term, though, these types of images are not allowed under Queue, so we have some breathing room.
Narrow-band filter detrend images will be a major challenge, and more experience is needed to do a good job here.
| |
|