Newsgroups: sci.geo.oceanography,sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.kei.com!bloom-beacon.mit.edu!eru.mt.luth.se!news.luth.se!sunic!ugle.unit.no!trane.uninett.no!eunet.no!nuug!statoil!swing!m28719
From: m28719@vega.statoil.no (Rune Aasgaard)
Subject: Re: What is the Helmert Transform?
In-Reply-To: bhattach@inf.ethz.ch's message of 1 Feb 1995 17:19:39 +0100
Message-ID: <M28719.95Feb6143140@vega.statoil.no>
Sender: usenet@statoil.no
Nntp-Posting-Host: vega.data.st.statoil.no
Organization: Statoil
References: <3g8gnj$r5g@orion.ethz.ch> <5epwVdeGaJB@alnilam.toppoint.de>
	<3gocar$32g@orion.ethz.ch>
Date: Mon, 6 Feb 1995 13:31:40 GMT
Lines: 27
Xref: glinda.oz.cs.cmu.edu sci.geo.oceanography:1028 sci.image.processing:12440

The Helmert transform is the surveying and geodesy name of the simple
2-dimensional linear conformal transform:

xt = a x - b y + dx
yt = b x + a y + dy

where a and b are :

a = S cos(fi)
b = S sin(fi)

Where S is the scaling factor and fi is the rotation angle.
dx and dy are the translations between the origins of the two
coordinate systems.

If you have defined a set of points in both the x,y and xt,yt
coordinate systems (2 or more) it is a simple process to compute a, b,
dx and dy by a least squares estimation. You should find the formulas
and algorithms in any basic text on land surveying, geodesy or
photogrammetry. 

If you don't find it in your library (and is unable to derive the
formulas yourself) I can dig up something more.
--
Rune Aasgaard, dr.ing. 			Tel    : +47 51805853
Statoil      				Fax    : +47 51805670
N-4035 Stavanger, Norway		E-mail : raas@statoil.no
