seeking info on tree disc analysis

• To: forest@nic.funet.fi
• Subject: seeking info on tree disc analysis
• From: mark@CS.MsState.Edu
• Date: Thu, 28 Jul 1994 18:17:26 +0300

I am a graduate student in Computer Science, and am working on a project that
is trying to analyze (count, measure, etc.) growth rings by using image
processing techniques on an image of the disc that has been scanned into a 
computer.

I recently saw a TV show where a person had a disc on a machine, and while 
looking through a "microscope" he hit a button when he saw the edge of a ring.
The machine was motorized to move the disc past the microscope, and I guess 
that it measured the distance between pushes of the button.  I do not know 
if he started from the center and worked his way out or started from one edge, 
worked his way to the center, and then out to the other edge. (does it matter?)
Is this the way everybody does it?  It sounds like a computer might spped 
things up quite a bit...

I am looking for "basic" information on how discs are analyzed - I have no
backgroung in forestry. 

Along with some "stupid" questions that I have, I am also interested in 
pointers to electronic copies of articles or other information that might be 
helpful.  ( I need to go to the bookstore to see if I can find a book or two)

Now for the stupid questions:

If you start at the center, and work your way to the edge, how do you decide 
which "line" you are going to follow?  (I have an image of a tree, and the 
tree looks like ot has been "flattened" at the bottom of the image - the
rings are closer together at the bottom than they are at the top or the sides.
Should I choose the side farthest from the "flattening"?)

Just on general principals, I would think that working your way from one edge,
through the center, and out to the other edge would allow you to compare the 
two sides, and possibly average them to get a "better" measurement.
Is this done?  If not, why not?

Do some researchers only care about the distance between the rings (on this 
one line) while others want to know the area of the rings?
For those who want the area, do they just take the measurements along this
line, and use these measurements to "build" a picture of a perfectly circular
tree (just calculating areas using the "points" along this line as the radius 
of a set of concentric imaginary circles?)

I am not a subscriber to this mailing-list (I found it through a Gopher server)
so please reply to me, not to the list.

Thanks
Mark Rauschkolb
mark@cs.msstate.edu

• Prev: Proceedings of the International Conference on Forest Vegetation Management
• Next: Info request
• Index: Forest mailing list August 1994
• Thread: Forest list subjects in August 1994