comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Cubic interpolation in 3D
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
Cubic interpolation in 3D [message #3346] Wed, 07 December 1994 19:57 Go to next message
rfulton is currently offline  rfulton
Messages: 5
Registered: November 1993
Junior Member
I am performing some 3D transformations on a 3D dataset using IDL 3.6.
The dataset is 64x64x64, which is actually a stack of contiguous
slices (each 64x64).

The T3D procedure is very nice for accumulating consecutive
translations and rotations in the 4x4 transformation matrix !P.T.

To apply the transformations I've tried the IDL Interpolate function, which
can be used for 1-, 2- or 3-D interpolation. Interpolate.pro allows
the use of the CUBIC keyword which gives the optimal interpolation for
my purposes. The trouble is, when I attempt 3D interpolation specifying
the /CUBIC keyword, the function appears to perform trilinear interpolation
rather than cubic. I seem to remember seeing somewhere in the documentation
that /CUBIC only applies to 1- or 2-D data, but I can't find it
again to confirm this.

Does anyone know for sure whether it is possible to perform 3D cubic
interpolation with the IDL Interpolate function ? If it is not,
is there any obvious reason why it has not been implemented. (eg.
does performing 3 sequence of three 2-d cubic interpolations give
exactly the same result as one 3-d cubic interpolation ? If that's
the case it would still be a good idea to allow 3D cubic interpolation
just to avoid extra coding IMHO.)

Alternatively has anyone modified or rewritten INterpolate.pro to allow
3D cubic ? Actually I'm interested in trying all kinds of interpolation that
may be implemented out there. Please let me know if you have any exotic
interpolation routines that you'd be willing to share.

Many thanks

Roger Fulton

===========================================================
Nuclear Medicine
Royal Prince Alfred Hospital
Sydney
Australia.

email: rfulton@atom.ansto.gov.au
tel: 61-2-516 8011
fax: 61-2-550 5172
===========================================================
Re: cubic interpolation [message #5227 is a reply to message #3346] Mon, 30 October 1995 00:00 Go to previous message
safier is currently offline  safier
Messages: 12
Registered: March 1995
Junior Member
>>>> > "brooker" == brooker <brooker@ccfm.ireq.ca> writes:

brooker> guys, Interpol from the user lib only does linear
brooker> interpolation of the input vector to an irregular grid
brooker> and I am affraid that I need to do a cubic
brooker> interpolation. Does anybody have this routine or must I
brooker> write it myself?

brooker> Thanks-peter brooker

brooker> p.s.- oui ou non? today is the day for canada

spline does a cubic spline interpolation. If you need cubic
interpolation in two dimensions I have translated the Numerical
Recipes bicubic spline routines to IDL, and will be glad to give them
to you.

Cheers,

Pedro
--
_/_/_/ _/ _/ _/_/_/ Pedro N. Safier
_/ _/ _/_/ _/ _/ Department of Astronomy
_/_/_/ _/ _/ _/ _/_/ U. of Maryland at College Park
_/ _/ _/_/ _/ phone: 301-405-1531
_/ _/ _/ _/_/_/ fax: 301-314-9067
  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: Multiple plots
Next Topic: Re: MESA

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Wed Oct 08 15:36:22 PDT 2025

Total time taken to generate the page: 0.00487 seconds