Here's what I'd do if I were in your position:
I'd spend 'time' working through both software manuals, and do simple
experiments with the software while doing so. I'd start by inputting
simple 'stick figures' to the software.
Draw a baseline 'stick figure' on paper:
Then make photcopies of this baseline 'image'. Work only with the
photocopies from here on out [because the photocopying process alters
the dimensions of the image [shrinks it a bit]].
Next, take one of your photocopies and add a line to it.
Then submit images 0 and 1 to the software of your choice, and do not
proceed until you've learned what you have to do manually to get the
simple stick figures to register perfectly.
Then check your learning of the software's needs by adding more
complexity to the stick-figure image. Work up to increased
complexity, through a number of trials, until you see that you can
'fool' the software by 'confusing' it [will probably be because one
subsection of your more-complex stick figure looks like another
section of your stick figure. Understand the software's 'confusion'.
Learn how to assist the software by manually setting "landmark"
points in your images, and do not proceed until you've learned what
you have to do manually to get the these more-complex stick figures
to register perfectly.
Then, take what you've learned and apply it to your dendritic images.
=Don't 'panic'= - there's no way around having to learn the needs of
the software, and how to take care of them through the software's
interface, so make the 'time' for it, and just do it. You only have
to go through this once, but what you learn will be useful to you
forever thereafter - so it's 'time' well-spent.
"Nick" <lifanshi at usc.edu> wrote in message
news:3e5e996b.12903163 at news.usc.edu...
| I have encountered the problems in image registration and
| and I need your help! Please see the detail in my website: