Re: [APML] AIP4WIN

[Roland Roberts <roland@astrofoto.org>]

>>>>> "ak" == Koebel, Alen <Alen.Koebel@christiedigital.com> writes:

    ak> How useful is AIP4WIN (which comes with the "Handbook of
    ak> Astronomical Image Processing" from Willmann-Bell) for
    ak> processing astrophotos scanned from film?  Does it do anything
    ak> essential (or even highly useful) that can't be done with
    ak> Picture Window Pro?

I don't find it all that useful for film photos.  There are some
things it can do that I can't do in PWP, e.g., a median filter is
useful as an alternate to a Gaussian blur for gradient removal; I
would argue it is *better* than a Gaussian blur but it is also a whole
lot slower.  

AIP4WIN will only work up to 16-bit FITS, and everything has to be
FITS.  It can import a lot of different formats, but not 48-bit TIFF.
I asked about that and Richard Berry told me it was not included due
to licensing costs for the TIFF library.  So you would have to convert
your images to some other format.  I can (and have) converted them to
"portable bitmap" (PBM) format under Linux and that allows using
AIP4WIN's generic importer.

However, I find that for most of my film shots, PWP is quick and does
a good job.  There is a learning curve and there are always judgement
calls in the processing where you can second guess yourself, but
AIP4WIN doesn't change that.

Also, there is/was a hard limit of 4k x 4k on the image size.  Richard
Berry told me he would consider increasing it but said things get
messy in Windows at the 4k point.  I don't do Windows :-) so I can't
comment.  For 35mm that's fine; you are unlikely to want to end up
with much more than about 3k x 2k pixels for 35mm.  At that point you
are already oversampling by about a factor of 2---the typical figure
of merit for a photographic lens is a 25 µm spot size.  For medium
format, you may want more pixels, especially if you plan to make a
large print.

    ak> I understand AIP4WIN can do image deconvolution (DDP or
    ak> Lucy-Richardson or ?).  Is deconvolution a useful tool for
    ak> processing images from film?  I have read that it can, within
    ak> limits, compensate for less-than-perfect tracking, among other
    ak> things.

I have taken one trailed image of Delphinus (which I think may have
been due to mirror flop in my guidescope---something I completely
forgot about when setting up to use a Mak-Cas as a guidescope) where I
used IRIS to to a maximum-entropy deconvolution and salvage something.
However, AIP4WIN doesn't include that particular trick.

Mind you, I do think AIP4WIN is a good tool, and if you are doing CCD
work it may even be a great tool.  I actually bought it primarily for
the book which I immediately read cover-to-cover (well, okay, I
skipped a few parts).  I think if you were going to do CCD work, I
would wholeheartedly recommend AIP4WIN as an economical, solid
package.  For film work I would consider it only so-so.

roland
-- 
		       PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD                             RL Enterprises
roland@rlenter.com                     76-15 113th Street, Apt 3B
roland@astrofoto.org                       Forest Hills, NY 11375


--  APML Archives at <http://astro.umsystem.edu/apml/>  ---
             Unsubscribe at <majordomo@seds.org>