[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Fwd: fits - Need to add CDELTn keyword.



jamie and all,

jamie is trying to figure out how the code to do all this processing will
get written or if it is written.  You experts please correct me if I am
wrong.  Here is my view of things:

Let me define two things:

Programs

Big masses of code that do big complicated things.  Like a Fortran compiler
or IRAF.

Scripts

Lists of commands that rout data to Programs.  Like the old DOS batch file.

Most of the programs that we need are written by big university groups that
have to do this sort of thing.  What is not written for the Mark IV data is
the sctipts.  

These do thing like:

1) take the evenings run that you find in this directory and:
a)send it to a program to dark and flat correct it
b) send the result to a program to find stars in CCD space
c)send the star list to a program to match these stars with a catalog and
label the stars in sky space.
d) take two exposures of the same part of the sky and keep only the
duplicate stars (thus throwing away the moving objects, and the cosmic
rays, etc.. 

...

Thsese scripts are relatively simple in form.  They are just lists of
commands.  Most of the time will be spent testing the scripts with data.
Finding a set of operations that gets the best results.  Fooling with
things like the aperture size used and the way a PSF function is developed.
 Trying different schemes to generate the best flat field.  That is why I
have been trying to get data out to the programmers so they can start
writing these scripts.  That is also why I am worrying about the .fts
header.  If it does not contain all the right information, then the scrips
will have to have manual entry of the data.  This will be a real pain.  

I urge all of you out there to look at disk 15, think about what needs to
be in the header, and someone work with me on RawToFits so that it gets
there.   

Hope this helps.

Tom Droege


At 10:11 AM 5/4/00 -0500, you wrote:
>
>tom
>
>> First, we are not yet reducing the Mark IV data.  But we are looking at
it. 
>
>there was my first erroneous assumption :[[[[
>
>> I am using Image Scientist.  People like Michael use fancier sturr.  You
>> take an image and then go to PSF.  You can auto generate a PSF.  Now you
>> can use the cursor to select an unsaturated star.  This leave you mag 9s
>> that you can find on Uranometria and then you can use the "Set Magnitude"
>> function of IS.  Now you can select "Create Star List".  This makes a star
>> list in x y coordinates.  If you previously select a few stars in the
>> "Astrometry" section so it can compute sky coordinates, then the star list
>> will have RA and Declination labels.  
>
>yes, i finally got to that point after much tedium . . can you imagine
>having to do that with every file on the disc, plus all the other discs of
>one night ?  impossible  . . there has to be batch way of handling all
>this so get your programmers to work, after all, you are paying them
>exhorbitant fees for this  . . .
>
>now, you got a list of coordinates of stars that are in the one frame . .
>no where does it inform this humble soul how to get it to work with the
>USNO-SAO.2.0 . . . and then to use it with tenxcat to get the range in
>magnitudes and compare these with past magnitudes to get light curves . . 
>if this is not yet possible to do with the present programs then i am
>whistling dixie . .  what programs are michael, arne, mike, etc. using to
>do all this at the present time ????
>
>by the way, if you want a good book to learn linux then look into sams
>teach yourself linux in 24 hours . . . very good, and it contains the
>cdrom for redhat
>
>take care, and thanks
>
>jamie
>
>
>