Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D

Notice
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
<-- Date Index --> <-- Thread Index --> [Author Index]

T3D Re: stereo & computers


  • From: "David A. Tiede" <D.A.Tiede@xxxxxxxxxxxxxxx>
  • Subject: T3D Re: stereo & computers
  • Date: Mon, 13 Apr 1998 15:56:53 -0500

john bercovitz wrote:
> 
> Is it OK to ask a question on this list if it's for my job?  8-)
> 
> I've got a network of ropes attached to the inside of a thin-
> walled pressure vessel and to each other.  The ropes meet at what
> I would call nodes some of which are on the vessel walls but most
> of which are in space.  I would name the nodes A, B, C,....N.  The
> connections between nodes I would name AB, AC, AF, etc.  There are
> a ton of nodes and the structure already exists.  What if I took
> pictures of it from more than one vantage point?  Is there a
> computer program smart enough to give me the positions of the
> nodes and the names of the connections from those pictures?
> 
> If you don't personally know of such a program but have a
> suggestion or two as to where I might start my search, that would
> be good info too.
> 
> Thanks,
> John B
I have written a program in the past using 4 cameras aimed down at the
jets of diesel fuel spray exiting from an injector nozzle.  It was
possible to determine the longitude and latitude of the various spray
lines.  The problem was aided by the knowledge that all the lines should
form a cone but the starting orientation was random.  The goal was to
extrapolate the lines to determine the closest distance to a central
known axis of rotation.  The idea is that all the spray lines should
intersect the central axis.  If they do not, the machinery which makes
the injector holes is adjusted until they do.

The program was written using a machine vision language called MacRail
and ran on a Mac.  The language was rather similar to Basic but came
with many specialized machine vision commands.  For example, the program
can define regions of interest in the image (which is acquired by frame
grabbers driven by the software as well).  Within these regions of
interest the program can find lines and determine their equations.  The
entire system was first calibrated using a calibration artifact which
consisted of a series of small spheres placed on posts of various
heights.  The program can find these various circles in each of the
views and helps determine the necessary rotation and translation and
scale matrix values.

The company supplying this system and software was called Automatix.  It
has been several years since I did this project.  I believe the company
was either bought out or changed it's name.  I remember the name Acuity.
The last address I have in my files is:

603 598-8400
Midwest Regional Office
8041 Hosbrook #105
Cincinnati, OH 45236
513 745-0653 513 543-3952 CAR
513 745-9688 FAX

This was a sales office.  The company was located in Billerica,
Massachuset.  I don't know if this information is up-to-date.

As far as starting a computer search, "machine vision" is certainly a
key phrase.  These types of programs are often used to control robots to
perform actions or to examine finished product to make sure everything
is present.

Incidentally, the final output of my program included a stereo pair of
the spray lines and the central axis so that the operator could have a
good idea what was going on with the geometry.


------------------------------