🚀 go-pugleaf

Article 3085 of sci.geo.meteorology:
Xref: ux1.cso.uiuc.edu sci.geo.geology:1951 sci.geo.meteorology:3085
Path: ux1.cso.uiuc.edu!uwm.edu!linac!att!bu.edu!dartvax!Timothy.Allen
From: Timothy.Allen@dartmouth.edu (Timothy Allen)
Newsgroups: sci.geo.geology,sci.geo.meteorology
Subject: SUMMARY (part 2/2) Image Processing Software
Message-ID: <1992Apr23.192619.23083@dartvax.dartmouth.edu>
Date: 23 Apr 92 19:26:19 GMT
Sender: news@dartvax.dartmouth.edu (The News Manager)
Organization: Dartmouth College, Hanover, NH
Lines: 594

SUMMARY OF IMAGE PROCESSING SOFTWARE, PART 2 of 2

Part 1 contains information about applications for (1) Macintosh
computers, (2) PC computers, (3) GRASS, and (4) Khoros.

Part 2 contains information about (5) other image processing 
packages for UNIX systems.
 

About two weeks ago, I posted the following query to several UseNet 
newsgroups and listserv mailing lists:

"I am interested in learning about what Image Processing software 
packages people are using for the analysis of multi-band data sets, 
such as Landsat Imagery. What are the major software packages in use? 
What are the capabilities of each package -- e.g. rectification, 
classification, PC and IHS transformations, filtering, contrast 
enhancement, overlays? Is the software designed specifically for 
remote-sensing images, medical images, or more generally for any kind 
of multi-band images? What are the hardware requirements? Any special 
requirements for display systems?"

"I am particularly interested in any software for Macintosh 
platforms, but would like to hear about packages that people are 
using on PCs and UNIX workstations as well. I am familiar with 
Terra-Mar's "MicroImage" software, and have heard of "ERDAS" but know 
nothing about it. I am also familiar with some single-band image 
processing software such as the excellent public-domain "NIH Image" 
program for the Macintosh."


Below is a summary of information that I have gathered so far.  I 
apologize for any mistakes or misquotes resulting from my butchering 
-- I mean my editing -- of the responses I received.  I welcome 
comments, corrections and further contributions to this list.

-------------------------------------------------------------------- 
Tim Allen                                    tim.allen@dartmouth.edu 
Earth Science Department, Dartmouth College
Hanover, NH 03755 
--------------------------------------------------------------------




--------------------------------------------------------------------
5 - Other Image Processing Software for UNIX systems
--------------------------------------------------------------------

apE 
WIT 
LaboImage 
Global Imaging Software 
IDL (IDL/PV-WAVE and IDL/SIPS)
SPAM 
IMDISP
IRAF 
AIPS 
HIPS 
RAND-CAGIS 
MicroStation Imager 
DISIMP
PCI 
BRL-CAD 

-----------------------------------------------------------------
From: "(Arnold Bloemer)" <bloemer@helios.tnt.uni-hannover.dbp.de>

apE - a huge visulization package with visual programming interface, 
in former times public domain, now free for universities (as I know). 
For information on availability of apE 2.x check at apE.osgp.osc.edu 
(128.146.18.18).

**now apparently available from:

TaraVisual Corporation 
929 Harrison Ave. 
Columbus, Ohio 43215 
1-(800)-458-8731   (614) 291-2912   FAX (614) 291-2867

----------------------------------------------------------------- 
From: "(Arnold Bloemer)" <bloemer@helios.tnt.uni-hannover.dbp.de>

WIT - commercial image processing package with a visual programming 
interface, a product of Logical Vision. Provided for use on the 
Catalyst CDware program as trial software; demo version on Sun CDware 
4.

Special hardware:	12 Mb. of memory and color monitor suggested 
Special software:	OpenWindows or X Window system

Logical Vision 
6882 Rupert St. 
Vancouver, B.C., Canada, V5S 2Z6 
Tel: 604-435-2587   Fax: 604-299-8263

-------------------------------------------------------------------- 
[forwarded by bloemer@tnt.uni-hannover.dbp.de <Arnold Bloemer>]

LaboImage:  COMPUTER VISION GROUP, Computing Science Center, 
University of Geneva

We are distributing (free) LaboImage, an image analysis software that 
we have developed during the past 3 years. It is used for teaching 
and research. It has a menu-driven interface, an extensive set of 
image processing/analysis capabilities (filterings, transforms, 
segmentation, binary and grey-level morphology, etc.), some special 
purpose tools (particles counting, 1D gel analysis, etc.). It is 
written in C. It currently runs under SunView; a Motif-X11 
implementation is planned. Interested people should contact:

Prof. Thierry Pun, Computer Vision Group 
Computing Science Center, U-Geneva 12, rue du Lac, CH-1207 Geneva 
SWITZERLAND  
Phone : +41(22) 787 65 82; fax: +41(22) 735 39 05 
E-mail: pun@cui.unige.ch [or pun@cgeuge51.bitnet]

------------------------------------------------------------ 
From: snow@netcom.com (Tim Szeliga)

We use Global Imaging Software to process AVHRR data, from the dish 
to the final display. Select a chunk of five band data from a pass, 
automatic navigation, calibrate it to Albedo and Temp, convert that 
to byte, register it to predesigned window, all relatively 
automatically and carefree.

It has no classification routines to speak of, but it isn't that 
difficult to write your own with their programmer's module.

Very small operation: one designs, one codes, one sells. Been around 
for a number of years, sold to Weather Service and Navy.  Runs on 
HP9000 with HP-UX.  Supports 24-bit display.

-------------------------------------------------------------------- 
From: Eric Schmidt <schmidt@sunset.larc.nasa.gov>

The IDL/PV-WAVE software package is what I currently use (the best 
available)...I/O is very simple (a major plus!!) & most of my 
programming is very short & sweet.  Numerous output formats are 
available.  there are certainly some minuses & it does require 
"programming".  I use the command line version, though the "point & 
click" version is available...I have some acquaintances who use it on 
the Mac (I think via Mac-X).  Basically, there are a lot of built-in 
functions, though some manipulation is required.  IDL/PV-WAVE is able 
to work with any type of imagery. IDL runs on a unix system.

-------------------------------------------------------------------- 
From: alan@elroy.Jpl.Nasa.Gov (Alan S. Mazer)

This is all in the context of Unix workstations but with that 
disclaimer...

Back in 1985 JPL developed something called SPAM (Spectral Analysis 
Manager) which got a fair amount of use at the time.  That was 
designed for Airborne Imaging Spectrometer imagery (byte data, <= 256 
pixels across by <= 512 lines by <= 256 bands); a modified version 
has since been developed for AVIRIS (Airborne VIsual and InfraRed 
Imaging Spectrometer) which uses much larger images.  What I hear, 
though, is that a lot of people are using IDL with a package called 
SIPS. This was developed at the University of Colorado (Boulder) by 
some people working for Alex Goetz (pronounced getz).  You might try 
contacting them if you already have IDL or would be willing to buy 
it.  It's a few thousand dollars (American) I expect for IDL and the 
other should be free.  Those are the general purpose packages I've 
heard of, besides what TerraMar has.

Spam does none of these things (rectification, classification, PC and 
IHS transformations, filtering, contrast enhancement, overlays).  
Actually, it does limited filtering and contrast enhancement 
(stretching).  I'm not sure about IDL/SIPS although I suspect IDL 
does a _lot_ more.  Spam is aimed at spectral identification and 
clustering.  SIPS _was_ written for AVIRIS imagery.  I'm not sure how 
general purpose it is.  You would have to contact Goetz or one of his 
people and ask.  I have another piece of software (PCW) that does PC 
and Walsh transformations with pseudocoloring and clustering and 
limited image modification (you can compute an image using selected 
components).  I've used it on 70 megabyte AVIRIS images without 
problems, but for the best speed you need an external DSP card.  It 
will work without it, but large images take quite a while (50-70 
times as long) to process.  That's a freebie if you want it.

The original Spam uses X or SunView to display.  The AVIRIS version 
may require VICAR, an executive based on TAE, and may also require a 
frame buffer.  I can refer you to people if you're interested.  PCW 
requires X for display.

--------------------------------------------------------------------
From a summary by ritley@uimrl7.mrl.uiuc.edu:

IMDISP Written at JPL and other NASA sites. It is available by FTP from
ames.arc.nasa.gov (128.102.18.3) in directory /pub/SPACE/SOFTWARE. Use
it
for simple display, enhancing, smoothing and so on. It is designed for
FITS
and VICAR/PDS images (the latter being NASA's planetary image format -
try
it out on images from the same site in the pub/SPACE/VICAR directory).
It
will also display TIFF files from any scanner, as long as you know the
size
of the image (lines and samples).

--------------------------------------------------------------------
From: rph63406@cs.fit.edu (Jose Suarez /ADVISOR Rafert)

The granddaddy of them all is IRAF (available from noao.edu or 
iraf.noao.edu), but it requires Unix, a lot of disk space, and a hell 
of a lot of patience. It's got extensive programming and I/O 
capabilities, and lots of packages have already been written for 
number-crunching within it. 

My favorite is IDL (Interactive Data Language) from Research Systems, 
Inc. (Oh, IRAF stands for Image Reduction and Analysis Facility.) IDL 
is in my opinion, much better and infinitely easier. Its programming 
language is very strong and easy -- very Pascal-like. It handles the 
number-crunching very well, also.

If you have A/UX, you can run IRAF. If you have money, you can run 
IDL (the Mac version isn't out yet, though -- I've only used it on 
Suns and VAX/VMS -- but it's coming.)

Personally, I like doing the number-crunching with IDL on the VAX (or 
Math- ematica, Igor, or even Excel on the Mac if it's not too hairy), 
then bringing it over to NIH Image for the imaging part. I have yet 
to encounter any situation which that combination couldn't handle, 
and the speed and ease of use (compared to IRAF) was incredible. 

By the way, it's mostly astronomical image processing which I've been 
doing. This means image enhancement, cleaning up bad lines/pixels, 
and some other traditional image processing routines. Then, for 
example, taking a graph of intensity versus position along a line I 
choose with the mouse, then doing a curve fit to that line (which I 
might do like in KaleidaGraph.)

--------------------------------------------------------------------
From a summary by ritley@uimrl7.mrl.uiuc.edu:

IRAF is free you can ftp it from noao.edu and complement it with STSDAS
from
stsci.edu. It is available via anonymous ftp from noao.edu.  Email to
iraf@noao.edu for more details.  Apparently this is one of the _de
facto_
standards in the astronomical image community.

-------------------------------------------------------------------- 
From: merlin%neuro.usc.edu@usc.edu (merlin)

IRAF -- contact iraf@noao.edu -- image reduction and analysis 
facility of the national optical astronomy observatory, kitt peak AZ 
-- software distributed by anonymous ftp from tucana.noao.edu -- also 
get a copy of saoimage for display under X windows -- extremely 
usefull package -- very flexible/extendable  -- tons (literally 3 
linear feet) of documentation for the general user, skilled user, and 
programmer

AIPS -- contact aipsmail@nrao.edu -- astronomical image processing 
system of the national radio astroinomy observatory, charlottesville 
VA -- software distributed by 9 track tape last time I ordered it -- 
I did not find this package very usefull -- tons of dense 
documentation -- very old code -- I heard in one of our astronomy 
laboratories the package may be rewritten completely in the near 
future -- major funding available.

-------------------------------------------------------------------- 
From: msl@cns.nyu.edu (Michael Landy)

I co-wrote and sell a general-purpose package for image processing 
which has been used for basically all the usual image processing 
applications (robotics, medical, satellite, engineering, oil 
exploration, etc.).  It is called HIPS, and deals with sequences of 
multiband images in the same way it deals with single images.  It has 
been growing since we first wrote it, both by additions from us as 
well as a huge user-contributed library. Here is a description of the 
software:

     HIPS  is  a  set  of  image  processing  modules  which together  
provide  a  powerful  suite  of  tools  for  those interested in 
research,  system  development  and  teaching. Originally  developed  
at  New  York  University,  HIPS  now represents one of the most 
extensive and flexible vision and image  processing environments 
currently available.  It runs under the UNIX operating system.  It is 
modular  and  flexi- ble, provides automatic documentation of its 
actions, and is almost entirely independent of special equipment.   
It  han- dles  sequences  of  images  (movies)  in precisely the same 
manner as single frames.  Programs and subroutines have been 
developed  for simple   image   transformations,  filtering, 
convolution,  Fourier  and other  transform processing, edge 
detection  and line  drawing   manipulation,  digital  image 
compression and transmission  methods, noise generation, and image   
statistics   computation.   Over   150   such  image transformation 
programs  have  been developed.  As a result, almost any  image 
processing  task  can be performed quickly and conveniently.  
Additionally, HIPS allows users to easily integrate  their  own  
custom  routines.   New users  become effective using HIPS on their 
first day.

     HIPS features images that are  self-documenting.   Each image  
stored  in  the  system  contains  a  history  of the transformations 
that have been applied to that image.   HIPS includes  a  small  set 
of subroutines which primarily deals with a standardized  image  
sequence  header,  and  a  large library  of  image  transformation 
tools in the form of UNIX ``filters''.  It comes complete with  
source  code,  on-line manual pages, and on-line documentation.  HIPS 
is now in use on a variety of computers including Vax and  Microvax,  
Sun, Apollo,  Masscomp,  NCR Tower, Iris, IBM AT, etc.  For image 
display and input, drivers are supplied for the Grinnell and Adage  
(Ikonas) image processors, and the Sun-2, Sun-3, Sun- 4, and Sun-386i 
consoles.  We also  supply  user-contributed drivers  for  a  number  
of  other framestores and windowing packages (Sun gfx, Sun console, 
Matrox VIP-1024, ITI IP-512, Lexidata,  Macintosh II, X windowing 
system, and Iris).  The Hipsaddon package includes an interface  for  
the  CRS-4000. It  is  a  simple matter to interface HIPS with other 
frame- stores, and we can put interested users in touch with  users 
who  have interfaced HIPS with the Arlunya and Datacube Max- Video.  
HIPS can be easily adapted for other  image  display devices because 
98% of HIPS is machine independent.

     HIPS has proven itself a highly flexible  system,  both as  an  
interactive  research tool, and for more production- oriented tasks.  
It is both easy to use, and quickly adapted and extended to new uses.  
HIPS is supplied on magnetic tape in UNIX tar format (either reel-
to-reel or  Sun  cartridge), and comes with source code, libraries, a 
library of convolu- tion masks, and on-line documentation and manual 
pages.

HIPS  stands   for   the   Human   Information   Processing 
Laboratory's Image Processing System.

     HIPS has been described in Computer  Vision,  Graphics, and  
Image  Processing (Vol.  25, 1984, pp. 331-347), and in Behavior 
Research Methods,  Instrumentation,  and  Computers (Vol. 16, 1984, 
pp. 199-216).

Michael Landy
SharpImage Software
P.O. Box 373, Prince Street Station
New York, NY   10012-0007
Voice:  (212) 998-7857
Fax: (212) 995-4011 (c/o Michael Landy, New York University)
Email:  landy@nyu.edu

-------------------------------------------------------------------- 
From: Robert Weissler <weissler%rondo@rand.org>

The RAND Corporation uses a package called RAND-CAGIS or simply CAGIS 
(Cartographic Analysis Geographic Information System) developed in-
house principally by Al Zobrist.  CAGIS is a modular GIS capable of 
raster, vector, and tabular data manipulation developed to support 
computer modeling and policy analysis research.  It uses a graphical 
user interface framework based on TAE Plus (Transportable 
Applications Executive, latest version 5.1) from Nasa-Goddard 
distributed by COSMIC.

CAGIS is organized as a toolkit for developing images and building 
applications around them.  It is used to prepare geographic data for 
computer models, facilitate smooth exchange of data between models, 
and perform supplementary calculations for these models. CAGIS 
includes a library of application programs for accessing and 
manipulating geographic databases, including DMA DTED and DFAD, World 
Database II, ETOPO5, SPOT and Landsat.  There are also other routines 
for image processing, converting formats, performing transformations, 
and providing analysis, as well as hooks to the Ingres relational 
database management system.  CAGIS can register geographic features 
to terrain and mosaic multiple image sources into a single image.

CAGIS has a unique compressed image format that all these routines 
can operate upon without uncompressing the image, saving roughly an 
order of magnitude disk space and considerable time over uncompressed 
images.  The compression technique allows the routines to index 
through the images in their compressed form to extract relevant 
portions quickly.

CAGIS runs on color Sun SPARCSTations (SPARC 2 GX recommended) 
running under SunOS 4.1.1, the X Window System and OpenWindows. It 
would probably benefit from running on a 2 GS where there is hardware 
support for Z and double buffering and hidden-line, hidden-surface 
removal.

We typically store our various geographic and other databases out on 
a juke box of eraseable opticals that can hold around 40 GB or more.

--------------------------------------------------------------------
From: lloyds@csis.dit.csiro.au

DISIMP (Device Independent Software for Image Processing) is a 
powerful system providing both user friendliness and high 
functionality in interactive times.

DISIMP incorporates a rich library of image processing utilities and 
spatial data options. All functions can be easily accessed via the 
DISIMP executive. This menu is modular in design and groups image 
processes by their function. Such a logical structure means that 
complicated processes are simply a progression through a series of 
modules.

Processes include image rectification, classification (unsupervised 
and supervised), intensity transformations, three dimensional display 
and Principal Component Analysis. DISIMP also supports the more 
simple and effective enhancement techniques of filtering, band 
subtraction and ratioing.

Running on UNIX workstations, DISIMP is capable of processing the 
more computational intensive techniques in interactive processing 
times. DISIMP is available in both Runtime and Programmer's  
environments. Using the Programmers environment, utilities can be 
developed for specific applications programs.

Graphics are governed by an icon-based Display Panel which allows 
quick enhancments of a displayed image. Manipulations of Look Up 
Tables, colour stretches, changes to histograms, zooming and panning 
can be interactively driven through this control.

A range of geographic projections enables DISIMP to integrate data of 
image, graphic and textual types. Images can be rectified by a number 
of coordinate systems, providing the true geographic knowledge 
essential for ground truthing. Overlays of grids, text and vector 
data can be added to further enhance referenced imagery.

The system is a flexible package allowing users of various skill 
levels to determine their own working environment, including the 
amount of help required. DISIMP comes fully configured with no 
optional extras. The purchase price includes all functionality 
required for professional processing of remote sensed data.

For further information, please contact:

The Business Manager, CLOUGH Engineering Group Systems Division, 627 
Chapel Street, South Yarra, Australia 3141. Telephone:  +61 3 825 
5555       Fax:  +61 3 826 6463

Email:  clough@csis.dit.csiro.au

--------------------------------------------------------------------
From: danw@ee.su.oz.au (Dan Wuescher)

I have just left a company called Intergraph (based in Huntsville  
Alabama) that sells a wide range of GIS software/hardware.  The 
product  I was working on was called Microstation Imager.  
Microstation is a base  graphics package that Imager sits on top of.  
Imager is basically an  image processing package with a heavy 
GIS/remote sensing flavor.  Some  of the capabilities that I can 
remember off the top of my head:

Basic geometry manipulations: flip, mirror, rotate, generalized 
affine.  Rectification: Affine, 2nd, 3rd, 4th and 5th order models as 
well as a projective model (warp an image to a vector map or to 
another image).  RGB to IHS and IHS to RGB conversion.  Principal 
component analysis.  Classification: K-means and isodata.  Fourier 
Xforms: Forward, filtering and reverse.  Filters: High pass, low 
pass, edge enhancing, median, generic.  Complex Histogram/Contrast 
control.  Layer Controller: manages up to 64 images at a time -- user 
can extract single bands from a 3 band image or create color images 
by combining various individual bands, etc.

There's probably a lot more I'm forgetting.  This would probably be  
a pretty high end solution ($ wise) but if you need something high  
end I recommend it.  The package is designed for a remote sensing  
application (it can handle VERY LARGE images) and there is all kinds  
of other software available for GIS applications.  

At the time of my departure it only ran on  Intergraph Workstations 
(a Unix machine similar to a Sun) though there  were rumors (there 
are always rumors) that the software would be  ported to PC and 
possibly a Sun environment.

--------------------------------------------------------------------
From:  Chris_F_Chiesa@cup.portal.com

  I'm a programmer first, graphic-arts programmer second, and remote- 
sensing integrator last, but I've encountered some software I can 
bring to your attention.  There's a copmpany called PCI, Inc., out of 
Richmond Hill (suburb of Toronto), Ontario, Canada, who makes an 
impressive array of software utilities for processing, manipulation, 
and use of remote sensing data in eight or ten different "industry 
standard" formats: LGSOWG, BSQ, LANDSAT, and a couple of others whose 
titles I forget.  The software is available in versions for MS-DOS, 
Unix workstations (among them HP, Sun, and IBM), and VMS, and quite 
possibly other platforms by now.  I use the VMS version.  

The "PCI software" consists of several classes/groups/packages of 
utilities, grouped by function but all operating on a common "PCI 
database" disk file.  The "Tape I/O" package is a set of utility 
programs which read from the various remote-sensing industry tape 
formats INTO, or write those formats out FROM, the "PCI database" 
file; this is the only package I use or know much about.  Other 
packages can display data from the PCI database to one or another of 
several PCI-supported third-party color displays, output numeric or 
bitmap representation of image data to an attached printer, e.g. an 
Epson-type dot-matrix graphics printer.  You might be more spe- 
cifically interested in the mathematical operations package: histo- 
gram and Fourier analysis, equalization, user-specified operations 
(e.g. "multiply channel 1 by 3, add channel 2, and store as channel 
5"), and God only knows what all else -- there's a LOT.  I don't have 
and don't use these, so can't say much about them; you only buy the 
packages your particular application/interest calls for. 

Each utility is controlled by from one to eight "parameters," read 
from a common "parameter file" which must be (in VMS anyway) in your 
"default directory."  (I found this annoying and patched the 
executable code to do it differently...)  Some utilities will share 
parameters and use the same parameter for a different purpose, so it 
can get a bit confusing setting up a series of operations.  The 
standard PCI environ- ment contains a scripting language very similar 
to IBM-PC BASIC, but which allows you to automate the process of 
setting up parameters for a common, complicated, lengthy or difficult 
series of utility execu- tions.  (In VMS I can also invoke utilities 
independently from a DCL command procedure.)  There's also an 
optional programming library which allows you to write compiled-
language programs which can interface with (read from/write to) the 
PCI data structures (database file, parameter file). My employers are 
too cheap to buy this official programming support, so I have had to 
scrounge a PCI manual and implement my OWN database/parameterfile 
reader/writer routines!

The PCI software is designed specifically for remote-sensing images, 
but requires such a level of operator expertise that, once you reach 
the level where you can handle r-s images, you can figure out ways to 
handle a few other thingds as well.  For instance, the Tape I/O 
package offers a utility for reading headerless multi-band (what 
Adobe PhotoShop on the Macintosh calls "raw") data from tape, in a 
number of different "interleave" orders. This turns out to be ideal 
for manipulating the graphic-arts industry's "CT2T" format, would 
probably (I haven't tried) handle Targa, and so on. Above all, 
however, you HAVE TO KNOW WHAT YOU'RE DOING or you can screw up to 
the Nth degree and have to start over.  It's worth noting that the 
PCI "database" file is designed to contain not only "raster" (image) 
data, but vectors (for overlaying map informa- tion entered via 
digitizing table), land-use, and all manner of other information (I 
observe that a remote-sensing image tape often contains all manner of 
information about the spectral bands, latitude/longitude /time/date 
etc. of the original satellite pass; all of this can go into the PCI 
"database").

I _believe_ that on workstations the built-in display is used.  On 
VAX systems OTHER than workstations PCI supports only a couple of 
specific third-party display systems (the name Gould/Deanza seems to 
come to mind).  One of MY personal workarounds was a display program 
which would display directly from a PCI "database" file to a Peritek 
VCT-Q (Q-bus 24-bit DirectColor) display subsystem.  PCI software 
COULD be "overkill" in your case; it seems designed for the very 
"high end" applications/users, i.e. those for whom a Mac/PC largely 
doesn't suffice (although as you know the gap is getting smaller all 
the time).  It's probably no coincidence that PCI is located in 
Canada, a country which does a LOT of its land/resource management 
via remote sensing; I believe the Canadian government uses PCI 
software for some of its work in these areas.

-------------------------------------------------------------------- 
From various sources:

The U. S. Army Ballistic Research Laboratory BRL-CAD Package is a 
powerful Constructive Solid Geometry (CSG) based solid modeling 
system.  BRL-CAD includes an interactive geometry editor, a ray 
tracing library, two ray-tracing based lighting models, a generic 
framebuffer library, a network-distributed image-processing and 
signal-processing capability, and a large collection of related tools 
and utilities. Release 4.0 is the latest version of software which 
has been undergoing continuous development since 1979.  It is 
available free of charge, and we encourage feedback and contributions 
from outside users.

We maintain an electronic mailing list CAD@BRL.MIL for general 
discussion, announcements, or bug reports related to BRL-CAD.  
Questions or requests for additions to this list can be sent to CAD-
REQUEST@BRL.MIL.

Informational files can be obtained by anonymous ftp from FTP.BRL.MIL 
(a.k.a. WOLF.BRL.MIL, 128.63.16.158), in the brl-cad directory.

Note that the files containing the actual software distribution are 
encrypted.  You will be sent a list hosts that contain the software, 
and details on how to decrypt the files, when you return the signed 
agreement (file "agreement").

BRL-CAD requires the UNIX operating system and is supported on more 
than a dozen product lines from workstations to supercomputers, 
including: Alliant FX/8 and FX/80, Alliant FX/2800, Apple Macintosh 
II, Convex C1, Cray-1, Cray X-MP, Cray Y-MP, Cray-2, Digital 
Equipment VAX, Gould/Encore PN 6000/9000, IBM RS/6000, Pyramid 9820, 
Silicon Graphics 3030, Silicon Graphics 4D ``Iris'', Sun Microsystems 
Sun-3, and the Sun Microsystems Sun-4 ``SparcStation''. Porting to 
other UNIX systems is very easy, and generally only takes a day or 
two.

For further details, call Ms. Carla Moyer at USA (410)-273-7794, send 
E-mail to <cad-dist@brl.mil>, FAX your letter to USA (410)-272-6763, 
or write to:

BRL-CAD Distribution 
SURVIAC Aberdeen Satellite Office
1003 Old Philadelphia Road, Suite 103 
Aberdeen MD  21001  USA

_____________________________________________________________________