| project_vol() in IDL [message #11174] |
Thu, 26 February 1998 00:00 |
gpetty
Messages: 7
Registered: June 1997
|
Junior Member |
|
|
Today's my first day using IDL and I must say I am very impressed with
its capabilities (BTW, David Fanning's book made it vastly easier for
me to take the plunge and actually accomplish something useful during
this foray -- thanks, David!)
However, I have already run into what seems to be an annoying
limitation in project_vol() and am wondering whether I'm just
overlooking something.
What I want to do is take a 3-D volume, and project it onto a 2-D
plane. OK, fine, that's what project_vol() does. But for the life of
me, I can't figure out a way to make the values of the projected image
consist of the CUMULATIVE TOTAL of the data values along a ray.
Rather, it seems to insist on doing an average or something similar.
(Same complaint applies to voxel_proj(), by the way).
To give a simple example, if I define a cubical array to contain data
values equal to 0.1 everywhere, project_vol() makes the cube look the
same shade everywhere, whereas it ought (for my application) to look
translucent, with the edges of the cube "thinner" (i.e., darker) than
the center.
Playing with OPAQUE etc. has not led me to a satisfactory solution
yet. DEPTH_Q seems to do something different, based on geometric
depth rather than cumulative data value.
Any suggestions for a workaround?
thanks
Grant
P.S. As long as I'm posting to this NG, I might as well mention one
other suggested improvement to IDL that immediately comes to mind:
command line history/editing/completion capabilities, a la the UNIX
tcsh shell. It's a pain to have to retype an entire lengthy command
from scratch when all I want to do is change one parameter!
--
Grant W. Petty |Assoc. Prof., Atmospheric Science
Dept. of Earth & Atmospheric Sciences |Voice: (765)-494-2544
Purdue University, 1397 CIVL Bldg. |Fax: (765)-496-1210
West Lafayette, IN 47907-1397, USA |Email: gpetty@purdue.edu
|
|
|
|
| Re: project_vol() in IDL [message #11176 is a reply to message #11174] |
Wed, 25 February 1998 00:00  |
davidf
Messages: 2866
Registered: September 1996
|
Senior Member |
|
|
Grant W. Petty (gpetty@rain.atms.purdue.edu) writes:
> Today's my first day using IDL and I must say...
Sigh...
My next book is going to be entitled How To Do Really, Really
Hard Things on Your First Day Using IDL. I'm going to use
examples I have collected from this newsgroup. :-)
> However, I have already run into what seems to be an annoying
> limitation in project_vol() and am wondering whether I'm just
> overlooking something.
I don't as a rule try to give advice without also attempting to
give some kind of an answer, but I've always believed rules were
made to be broken.
Here is my advice. Sounds like you want to do some fairly
complex 3D kinds of things. I would NOT be doing these in
direct graphics, which is what Project_Vol and Voxel_Proj
use. IDL in this incarnation is really a 2 and a half D
application. As long as you are committed to learning
something new, I would plunge in and try to learn how to
do this using the new object graphics, which were designed
*specifically* for this kind of 3D application.
This topic is NOT covered in my book because, frankly, I
haven't been able to figure out exactly how it works
from the documentation that is supplied with IDL. (I am
currently scratching my head over Ken Bowman's polygon
object problem. I can get *almost* there, but not quite,
and I end up doing "experiential programming" in which I
make random changes in my programs, hoping beyond hope
that something will start to make sense. I hate it.)
What I *have* found is that the people at RSI are wonderfully
helpful with giving me advice when I ask for it. And there
are clearly people there who understand how this object graphics
stuff works. I think if we all work together on this
that sooner or later we will be able to get together
some simple examples that demonstrate the principles. The
fact that there are two object graphics questions here in
two days certainly makes me think the time is right for *me*
to learn it better than I do now.
Just looking, for example, at the Volume Object I see that
it has four different "composite functions" for how to determine
the value of the projected pixel. One of these is an "alpha sum",
which sounds to me like what you want. There is also depth cueing
and various opacity and lighting options. I would pester the
folks at RSI until you learn how it works, and then publish a
good example here. :-)
The alternative is to look at the Project_Vol source code, which
is written in IDL and is available in the lib subdirectory, and
make the modifications to it yourself. This is probably not a
first-day-with-IDL kind of a job, but I didn't think the source
code looked impossibly hard, either.
> P.S. As long as I'm posting to this NG, I might as well mention one
> other suggested improvement to IDL that immediately comes to mind:
> command line history/editing/completion capabilities, a la the UNIX
> tcsh shell. It's a pain to have to retype an entire lengthy command
> from scratch when all I want to do is change one parameter!
Now here is a question I *DO* know how to answer. :-)
There should be no reason to re-type a lengthy command from
scratch. IDL has a command "history" buffer that is accessible
using the UP arrow key. Simply recalling the command and
editing it before hitting the Carriage Return will do the
job. The history buffer can be increased in size to more
than the default 20 commands if need be. (I was going to
give you a reference in my book, but I see that section
somehow got deleted in my overzealous editing. I'm going
to correct this in the next printing, even if it means
adding a couple of more pages.)
Sometimes the terminal window on UNIX machines has to be
configured properly to get command recall working. Let me
know if this is happening to you, and we can fix it.
Cheers,
David
-----------------------------------------------------------
David Fanning, Ph.D.
Fanning Software Consulting
E-Mail: davidf@dfanning.com
Phone: 970-221-0438
Coyote's Guide to IDL Programming: http://www.dfanning.com/
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
