Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!news.duke.edu!convex!cs.utexas.edu!howland.reston.ans.net!news.sprintlink.net!pcnet.com!circellar!carl.giles
From: carl.giles@circellar.com
Subject: ** PROJECT **
Message-ID: <9503131639.0NEUW02@circellar.com>
Organization: Circuit Cellar, Inc.
X-Mailer: TBBS/PIMP v3.25
Distribution: world
Date: Mon, 13 Mar 95 16:39:56 -0500
Lines: 32



    ____________________________________________
   |       . .                  . .             |
   |     .     .               .   .         . .|
Y  |    .        .     .      .     .     .     |
   |....         .    . .   .        .   .      |
   |              . .    .             .        |
   |____________________________________________|
                         X

Well, imagine that this is a simple function drawed on a paper. The length 
(X) of the paper is divided in 144 parts, one each 10 minutes. In each part, 
he writes the level of rain in that moment (my english is not very good :) 
That me that after one day, he has a register of the falling rain in that 
day.  Well, go ahead: He wants to put that levels in a file of his computer, 
wants an electronic system to read that lines/curves and convert them to
data.
  
  My suggestion is to get a HAND SCANNER to scan in a digitized image
of the paper (if the paper is less than 4 inches wide, otherwise use
a page scanner.)  Since color scanners are so popular now, a monochrome 
scanner should be dirt cheap!  Whatever graphics format you decide to 
use, get the information on the format, then write software to find the 
X and Y positions of the dots by comparing the positions to the borders 
of your chart.  The software would not be easy, but it could be written 
in a high level language on the PC.
 
    ___      _ 
 _--   --_ _-                    Regards,
-_       _-_                     Carl Giles
  --___--   -_                   ( carl.giles@circellar.com )
