Posted by
Benjamin Schmid-2 on
URL: http://imagej.273.s1.nabble.com/Understainding-how-3D-viewer-works-tp4999802p4999808.html
Hi,
On 22:03 Thu 16 Aug , Eugeniu Nacu wrote:
> Hi everyone,
>
> I am acquiring Z-stacks with the Zeiss LSM780 to acquire confocal
> images, and I save them in the LSM5 format. I want to reconstruct
> the stacks into a 3D image.
>
> Since I am new to using the 3D viewer and I was wondering if someone
> could explain how the 3D viewer actually works. I am interested in
> knowing what metadata is used and what actually is going on when the
> 3D image is created.
>
> Also, I was wondering if you could give me advice on the following issue:
> I have an undersampled sample such as the following: it is a
> 483.03µm thick z-stack, with 10 z-slices, z-slice thickness of 2µm,
> and z-step of 53.67µm.
> I want to create a 3D view of this sample where my 2µm z-slices
> would be rougly 53.67µm apart.
> Right now it seems that the Fiji just uses the z-step of 53.67µm as
> the voxel depth instead of 2µm. Also it seems to not know the
> difference between voxel depth and z-step because when I change the
> voxel depth to 2µm the 3D viewer "squashes" the image into 20µm by
> simply layering the different z-slices on top of each other.
> Any idea how to make the 3D viewer read the correct voxel depth and
> the z-step so that it can spread my z-slices accordingly in the 3D
> image?
>
> Thanks a lot!
> Eugen
The 3D Viewer indeed assumes that the z-step equals the slice thickness.
It reads this information from the image calibration, which can manually
be adjusted under -Image-Properties. Unfortunately, there's no way to
specify the z-step there, so no chance for the 3D Viewer to get that
info.
For a quick solution which doesn't require any programming, you could just
fill your stack with empty slices. I can also implement some function to
separate slices in x-, y- and z-dimensions in the 3D Viewer. However it
might take a few days before it's done.
Volume renderings in the 3D Viewer are done in the following way:
For each plane, a rectangle is created which gets a texture applied
(namely that particular plane). The distance between rectangles is the
voxel depth. Rectangles are created in x-, y- and z- direction, however
only one set is visible a time, depending on the view orientation.
>
> --
> Dr. Eugeniu Nacu
>
> Postdoctoral Fellow in Group of Prof. Dr. Elly Tanaka
> Center for Regenerative Therapies Dresden
>
> Fetscherstr 105, 01307 Dresden, Germany
>
> Tel: +49 351-458 82037
> Fax: +49 351-458 82009
Hey, you are just next door, I'm at the MPI-CBG, so we could also meet
if you want.
>
> I am currently trying to reduce the number of times I check my email per day in order to avoid distractions and increase productivity.
> Therefore my response to your emails may be somewhat slow.
>
me too ;)
Cheers,
Bene
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html