Neuron structure from photograph

[Aaron King]

Does anyone have information on processes by which neuron morphology information from photographs
and confocal light microscopy can be translated to data suitable for use in compartmental modeling
(i.e., lengths and diameters of cylindrical segments)?  Does specialized software exist for the
purpose?  If it does, is it well-developed software?  

Another approach to this same problem is to "grow" dendritic trees using an idealized branching 
process, such as the L-systems approach of Lindenmayer (1968).  This system uses production rules
to generate a tree structure, which can then be compared with the real thing, as recorded in photo-
graphs, for example.  Has anyone heard of this sort of an approach being tried before?

I'd really appreciate any help you can give me.  Thanks much,

Aaron King
aking at math.arizona.edu