-------- Original Message --------
Subject: Re: image taking for geometric morphometrics
Date: Tue, 6 Sep 2011 17:25:24 -0400
From: Brendan McCane <[email protected]>
To: <[email protected]>
If you want to take accurate measurements from images, then there is a
whole field on those techniques called photogrammetry. Much of that
field deals with how to get accurate 3D measurements from multiple
images of the same scene/object, but some of the techniques are also
important for 2D measurements. In particular it is important to
determine the internal parameters of the camera (focal length, lens
distortion, relative pixel size in x and y directions, centre of
projection etc). There is some software around for doing this and it is
relatively straight-forward once you know how, but still not quite
trivial (a matlab toolbox is here:
http://www.vision.caltech.edu/bouguetj/calib_doc/). Usually one needs to
take a picture of a calibration object of known size and position and
then relate landmarks from the calibration object back to landmarks in
the image.
However, all of that may be unnecessary if your setup is good enough,
but only you can determine that. As a minimum you should take some
images of objects of known size and geometry (checkerboard patterns are
good choices) and see if measurements on the image produce results
close-enough to real measurements for your purposes. If not, then you
will have to do some camera calibration.
On Sep 7, 2011, at 12:35 AM, morphmet wrote:
-------- Original Message --------
Subject: image taking for geometric morphometrics
Date: Mon, 5 Sep 2011 19:18:07 -0400
From: Darren Parsons <[email protected]>
To: <[email protected]>
Hi there,
I am just starting out with geometric morphometrics, so was hoping
someone could point me towards any documents that could get me started.
My specific query is around the image taking process. This is because I
may not be able to analyse the images for some time (images will be of
fish ~ 40 cm in length). So, for now, I would like to ensure that I take
images that will allow me to do 2D analysis later on. Is there a
document/book/previous string of postings that outlines the best image
taking process/equipment in simple terms? I have "Zelditch, Geometric
morphometrics for biologists : a primer" on order from my library?
Any help would be greatly appreciated.
Thanks,
Darren
(please send replies to [email protected] <mailto:[email protected]>)
Darren Parsons
National Institute of Water and Atmospheric Research (NIWA)
41 Market Place, Private Bag 99940, Auckland, New Zealand
ph: +64-9-375 4531
fax: +64-9-375 2051
cell: 021 170 1724
email: [email protected] <mailto:[email protected]>
Please consider the environment before printing this email.
NIWA is the trading name of the National Institute of Water &
Atmospheric Research Ltd.
Cheers,
Brendan.
-----------------------------------------------------------------------
Brendan McCane, Head of Dept, Email: [email protected]
Department of Computer Science Phone: +64 3 479 8588/8578.
University of Otago Fax: +64 3 479 8529
Box 56, Dunedin, New Zealand.
