comp.lang.idl-pvwave archive: archive » Matching, Aligning, Affine Transform

Home » Public Forums » archive » Matching, Aligning, Affine Transform

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Matching, Aligning, Affine Transform [message #27353]

Thu, 18 October 2001 05:09

azM
Messages: 8
Registered: October 2001

Junior Member

Has anybody written a routine to match two images or even better volumes? At
present I'm using a set of external programs (MIRIT, FSL-FLIRT both on UNIX)
to calculate translation and rotation of a float volume versus a reference
volume. The result is a 4x4 (affine) transformation matrix. (I posted some
questions about this matrix earlier). I'm thinking about writting the whole
routine myself because I'm not satisfied with the results of the UNIX
programs. Unfortunatly I probably don't have enough time for my internship
left to fully complete this. I want to base it on so called cost algorithems
(Woods functions, Joint Entropy etc). Can anybody help me?

Thanks in advance,
Bob
B.C.Hamans<<at>>student.tue.nl

Report message to a moderator

Re: Matching, Aligning, Affine Transform [message #27380 is a reply to message #27353]

Mon, 22 October 2001 13:06

gogosgogos
Messages: 15
Registered: July 2001

Junior Member

I run into this problem and realized the source code
available for IDL is minimal..

a few suggestions:

1. grab dll-to-lib programs and create static libraries and link them to idl
2. if you have to present the source code, go get the free C code
of AIR and SPM and translate that into IDL.. (stupid thing but if
your instructor is a dumbass, that should get you off the hook)..

golden rule: IDL is a nice program, but only if you accept
its limitations... one of them is big confusement when it comes
to 3d transforms and such... the new image processing toolbox
for matlab does this with a single command...

i will not upgrade to 5.5 for sure... it still is behind matlab..

flame me for my words, i like some activity in the chan =)

Report message to a moderator

Re: Matching, Aligning, Affine Transform [message #27466 is a reply to message #27380]

Tue, 23 October 2001 01:57

Nigel Wade
Messages: 286
Registered: March 1998

Senior Member

GB Smith wrote:

> golden rule: IDL is a nice program, but only if you accept
> its limitations... one of them is big confusement when it comes
> to 3d transforms and such... the new image processing toolbox
> for matlab does this with a single command...

But you are not comparing like products. The signal processing toolbox is
not a part of core MATLAB. It is an additionally priced product and you
have to pay extra maintenance to keep using it.

>
> i will not upgrade to 5.5 for sure... it still is behind matlab..

Really?
Please explain how the following demonstrates the superiority of MATLAB 6
over a quite old version of IDL:

< M A T L A B >
Copyright 1984-2000 The MathWorks, Inc.
Version 6.0.0.88 Release 12
Sep 21 2000

>> i=int16(1);
>> i=i+1;
??? Error using ==> +
Function '+' not defined for variables of class 'int16'.
>> i
i =
1

IDL Version 5.1.1 (IRIX mipseb). Research Systems, Inc.

IDL> i=fix(1)
IDL> i=i+1
IDL> print,i
2

To me, any product which can't do a simple 1+1 is not worth considering for
data processing.

>
> flame me for my words, i like some activity in the chan =)

Glad to oblige.

--
-----------------------------------------------------------
Nigel Wade, System Administrator, Space Plasma Physics Group,
University of Leicester, Leicester, LE1 7RH, UK
E-mail : nmw@ion.le.ac.uk
Phone : +44 (0)116 2523568, Fax : +44 (0)116 2523555

Report message to a moderator

Previous Topic:	IDLWAVE mode for emacs
Next Topic:	Read HTML Page?

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Wed Oct 08 19:39:10 PDT 2025

Total time taken to generate the page: 0.00639 seconds