-------- Original Message --------
Subject: Re: Software for standard Fourier analysis
Date: Thu, 7 Aug 2008 00:26:31 -0700 (PDT)
From: [EMAIL PROTECTED]
To: [email protected]
References: <[EMAIL PROTECTED]>


Hi, I would recommend against the method. The first problem is finding a
reasonable origin. The program searches for this iteratively, minimizing
the first harmonic, but it is a mess and sometimes converges on something
silly. Also, the constant-angle requirement means that the outline can
become very unevenly sampled (also raising interpolation issues). All this
makes FA a more complicated method to program than EFA! Finally, only the
simplest shapes can be analyzed at all, because radius must be a
single-valued function of angle.

Actually, the main reason I include it in Past is pedagogical - to
demonstrate its poor performance ...


Oyvind Hammer
Natural History Museum
University of Oslo


-------- Original Message --------
Subject: Software for standard Fourier analysis
Date: Wed, 6 Aug 2008 13:15:01 +1000
From: Paul Sanfilippo <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]

Dear Morphmet Community,

I'm in the early stages of some data collection on outlines of optic
discs.  I've been primarily using elliptic Fourier analysis for this,
but thought I should also do the analyses using standard Fourier
methods.  There's a plethora of software out there for EFA, but very
little for standard FA (I see PAST does it). Are there any such bits
of software still floating around out there?

Thank you,

Paul Sanfilippo
PhD Student
Uni Melbourne
Australia


--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org






--
Replies will be sent to the list.
For more information visit http://www.morphometrics.org

Reply via email to