Hi Bill , VisAD team and eveyone on the list,
I am currently extending JAMA to include all methods(functions) which are
not available from the
current version. I call it XJamlab (Extended Java Matrix Laboratory), and it
is basically writing Java
versions of functions found in MatLab. Currently , I have not documented it,
but scientists or engineers who have
been developing using MatLab can find methods in XJamlab exactly the same as
in MatLab and does
exactly the same matrix operations, like flipLR( ) , flip matrix
left-to-right, flipUD( ) , flip matrix upside-down, etc, etc.
This translation of MatLab to Java is not completed yet, but will be
sometime in the future. Since I do not have a
website of my own to put it up for download by interested developers,
anybody who wants XJamlab can contact me.
----- Original Message -----
From: "Bill Hibbard" <hibbard@xxxxxxxxxxxxxxxxx>
To: "Wilson, Andrew, J" <andrew.wilson@xxxxxxxxxxxx>
Sent: Tuesday, July 30, 2002 1:20 AM
Subject: Re: Cluster analysis, principal components,regression analysis etc
for image segmentation
> Hi Andrew,
> > Is there anyone else out there working on cluster analysis,
> > principal components etc, maybe for the purpose of image
> > segmentation, and would like to share code, test data sets
> > etc?
> The Java Matrix Package (JAMA) provides a variety of matrix
> operations useful for these analyses. For details see:
> VisAD provides a way to apply JAMA operations to VisAD Data
> objects via its visad.matrix package (this requires that you
> install JAMA, of course), as well as support for Fourier
> transform and histogram operations via the visad.math package.
> These operations are also accessible from Python. See:
> for a list (and see Tom's Python tutorial for more general
> information. The visad/python/matrix_test.py program is an
> example of a simple matrix solution using JAMA.
> Bill Hibbard, SSEC, 1225 W. Dayton St., Madison, WI 53706
> hibbard@xxxxxxxxxxxxxxxxx 608-263-4427 fax: 608-263-6738