| Re: mosaic routine? [message #64448 is a reply to message #64446] |
Fri, 26 December 2008 12:12  |
wlandsman
Messages: 743
Registered: June 2000
|
Senior Member |
|
|
On Dec 26, 11:01 am, Dick French <rfre...@wellesley.edu> wrote:
> Hi, folks -
> I'm trying to create an accurate mosaic composite image from about 45
> tiles. In the simplest instance, no rotations or reprojections are
> required - just sliding rectilinearly.
I presume these are astronomical images == can you get an astrometric
solution for each individual tile? This would automatically give the
shift between individual tiles. If I had astrometric information in
each individual FITS header, then I would probably create an empty big
image with a FITS header, and use hastrom.pro (http://
idlastro.gsfc.nasa.gov/ftp/pro/astrom/hastrom.pro ) to map each
individual tile into the big image. HASTROM uses POLY_2d
internally which gives the option of using bilinear or cubic
interpolation for subpixel shifts.
I'd be very surprised if you couldn't get an astrometric solution for
each tile, given the availability of huge astrometric reference
catalogs like USNO, and automatic plate solution software like
astrometry.net.
But if for some reason you need to correlate images to get relative
shifts, you might look at some very old IDL mosaic software written by
Frank Varosi ( http://adsabs.harvard.edu/abs/1993ASPC...52..393V ) for
small infrared arrays, which is still available at
http://idlastro.gsfc.nasa.gov/ftp/contrib/varosi/
Finally, you might look at the SIMPLE software
( http://www.aoc.nrao.edu/~whwang/idl/SIMPLE/MOIRCS/Doc/index. html)
which is designed to create a mosaic of dithered optical images, but
does much more than you probably need. --Wayne
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
