In article <dougC9nJL5.J5y at netcom.com> doug at netcom.com (Doug Merritt) writes:
If I now interpret your question correctly, you actually want something
to glue sections together into a 3D model, which would involve detection
of object boundaries versus object interiors in the slices, and then
fitting polygons or some kind of 2d curves to pass through those edges.
The resulting 3d models would then be displayed by some handy rendering
package.
We do this sort of thing in our lab and have produced pictures for the cover
of the Journal of Neuroscience several times. It is not an easy task and
requires the user to sit down and manually trace out contours using a trackball
or a mouse on digitized images. In addition, we use a blink comparator to
align the sections. The actual 3d reconstruction from the serial 2d contour
dataa is performed on a workstation running commercial 3d reconstruction
software.
For more information, send me email: davis at amy.tch.harvard.edu
--
_____________
#___/John E. Davis\_________________________________________________________
#
# internet: davis at amy.tch.harvard.edu
# bitnet: davis at ohstpy
# office: 617-735-6746
#