1
From glynn.clements@virgin.net Wed Dec 12 19:13:35 2001
2
Return-Path: <glynn.clements@virgin.net>
3
Received: from camelot.itc.it (camelot [195.223.171.5])
4
by artemide.itc.it (8.11.3/8.11.3) with ESMTP id fBCIDX728309
5
for <neteler@itc.it>; Wed, 12 Dec 2001 19:13:33 +0100 (MET)
6
Received: from mta2-svc.virgin.net (mta2-svc.virgin.net [62.253.164.42])
7
by camelot.itc.it (8.11.3/8.11.3) with ESMTP id fBCIDWd06840
8
for <neteler@itc.it>; Wed, 12 Dec 2001 19:13:32 +0100 (MET)
9
Received: from cerise.nosuchdomain.co.uk ([62.252.69.31])
10
by mta2-svc.virgin.net (InterMail vM.4.01.02.27 201-229-119-110)
12
id <20011212181330.PKYR10663.mta2-svc.virgin.net@cerise.nosuchdomain.co.uk>
13
for <neteler@itc.it>; Wed, 12 Dec 2001 18:13:30 +0000
14
Received: (from glynn@localhost)
15
by cerise.nosuchdomain.co.uk (8.11.6/8.11.6) id fBCIAC301072;
16
Wed, 12 Dec 2001 18:10:12 GMT
17
From: Glynn Clements <glynn.clements@virgin.net>
19
Content-Type: text/plain; charset=us-ascii
20
Content-Transfer-Encoding: 7bit
21
Message-ID: <15383.40324.300167.67864@cerise.nosuchdomain.co.uk>
22
Date: Wed, 12 Dec 2001 18:10:12 +0000
23
To: Markus Neteler <neteler@itc.it>
24
Subject: Re: affine transform
25
In-Reply-To: <20011212165039.B4313@itc.it>
26
References: <20011212165039.B4313@itc.it>
27
X-Mailer: VM 6.94 under 21.4 (patch 4) "Artificial Intelligence (candidate #1)" XEmacs Lucid
35
> by chance I have found another implementation for affine
36
> transform (or whatever there is implemented):
37
> src/libes/vect32/libes/transform/
38
> It is used in v.digit and maybe other programs.
40
> Is this a candidate for the GMATH library?
43
The code there seems fairly specialised.
45
The transformations are affine, but the means of generating the
46
coefficients seems rather odd; in particular, the requirement that at
47
least four reference points are provided. Three points will define an
48
affine transformation; any more are either redundant or contradictory.
50
If it were to go into the gmath library, it might be better for
51
compute_transformation_coef() to return the resulting transformation
52
in a gmath "mat_struct". transform_a_into_b() and transform_b_into_a()
53
are largely redundant; they're either just a matrix multiplication (if
54
using homogeneous coordinates) or multiply-and-add (otherwise).
56
Glynn Clements <glynn.clements@virgin.net>