Subject: | |
From: | |
Reply To: | |
Date: | Sun, 23 Oct 1994 22:03:41 -0700 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Lance Ladic writes:
>P.S. As a followup question, if VoxelView will not allow me to obtain a more
>accurate visual representation of the scale of my reconstructions in the
>Z-plane, is there any other volume rendering software that will?
Can't comment on VoxelView, but I have solved this very problem with AVS.
It is a simple matter of adjusting the relative scale of the axies
aplied to the ray tracer, using a 4x4 Euler transform. One can apply a
floating point coefficient to the transform elements and come out with a
...what is the word...correctly proportioned image. For example where
we use our 100x objective and have 512x512 images with x-y pixel sizes
of ~0.1micron square, we use a z-axis step size of 0.5micron. To
correctly render the image, we would simply multiply the axies by 1,1, and 5
in x,y, and z, respectively.
(Actually though, this is also a good place to partially correct for focal
anomalies introduced by mismatch in refractive indicies of immersion oil
and mounting media, wherein z-axis stage movement does not equal focal point
movement. The result being that your .5 micron stage movement turns out
to be somewhat less in the collected image.)
I was just last week trying to describe the nature of the problem, and I find
it interesting to see the way it is put. There must be some word to
describe the relation of correct dimensions. Something like "ortho-dimentional"
or "iso-lateral", anyone have a good word for it?
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Andy Jacobson <[log in to unmask]> <[log in to unmask]>
Dept. Pharmacology / Div. Nuclear Medicine and Biophysics
CHS B2-086
UCLA School of Medicine Phone:310-825-8584
Los Angeles, CA 90024-6948 Fax: 310-825-4517
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
|
|