No subject

[Net.Noise owner]

Newsgroups: info.grass.user
Path: zorro.cecer.army.mil!shapiro
From: shapiro at zorro.cecer.army.mil (Michael Shapiro)
Subject: Re: problems with i.rectify
Message-ID: <C1J2H2.CAF at news.cecer.army.mil>
Sender: news at news.cecer.army.mil (Net.Noise owner)
Organization: US Army Corps of Engineers Construction Engineering Research Labs
References: <9301131904.AA11778 at hanover-crrel.army.mil>
Date: Wed, 27 Jan 1993 19:36:37 GMT
Lines: 28

In <9301131904.AA11778 at hanover-crrel.army.mil> samg at hanover-crrel.army.mil (Sam M. Gustman) writes:

>     I wrote in before asking for help with i.rectify.  I have successfully
>used i.target and i.points.  I get an RMS error smaller then my pixel size.
>However, when I perform i.rectify my image gets blown up to enormous 
>proportions.  The region I am rectifying to fully contains my image.  Are 
>there any other suggestions as to what could be happening?
>             Thanks,  Sam Gustman...

I'm not sure from this description ("gets blown  up  to  enormous
proportions")  what the problem is. Is the image much larger than
the region? Does it have too many rows and cols  in  it?  Is  the
image distorted in some way?

I have seen odd behaviour from i.rectify if you don't pick enough
and properly spaced control points in i.points.  i.rectify uses a
linear transformation (affine) which requires  the  determination
of  6  parameters (3 for the x direction, 3 for the y direction).
If you do not specify enough control points, or  if  the  control
points  tend  to  lie  on  a  line,  then you can get a distorted
rectification. 4 points is a technical minimum,  but  you  should
use more (perhaps 10 or 15) and they should be spread out over as
much of the image as possible.
-- 


Michael Shapiro U.S. Army CERL
Environmental Division