-------- Original Message --------
Subject: Re: Next Engine scanner protocol
Date: Mon, 9 Apr 2012 14:09:32 -0400
From: Heather Garvin <[email protected]>
To: [email protected]
I haven't scanned anything that small(I was scanning human skulls),but I
can tell you that paying the extra money for the HD Pro software makes a
world of a difference. Not only does it give you higher quality scans,
but it cuts your scanning time pretty much in half.
Besides that, I was given advice by Matt Tocheri to use the maximum
number of divisions under the highest speed. This may sound
counter-intuitive, but the rationale is that you get better accuracy
when you have more regions of overlap (from different angles). With a
high degree of overlap you can obtain just as many surface points than
going more slowly over less divisions. Also in the slower speeds (higher
dpi), it was suggested to me because the laser is going over each region
more slowly, there's more of a chance it will pick up noise or reflect
back oddly, creating more inaccuracies. So I followed this advice with
human skulls (using the HD program, highest number of divisions, fastest
speed). I'm not sure how it would work out on the smaller skulls.
Finally, the NextEngine software does okay aligning the divisions from a
single scan (given that the object doesn't move). I use the "Volume
Merge" instead of any of the "fusion" options, to get a uniform mesh.
But I found that it really sucked at aligning more than one 360 degree
scan together (for example, if you took one 360 scan, changed position
and took another). So after each scan I would save the .ply and open it
in another program, such as GeoMagic or Rapidworks, and do the rest of
the aligning and merging in there. The problem is, these programs are
expensive, but if you have an associated engineering program they may
already have an institutional copy.
Hope this helps a little. Goodluck!
--Heather
*
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Heather Garvin*
PhD Candidate
Center for Functional Anatomy & Evolution
Johns Hopkins School of Medicine
http://www.hopkinsmedicine.org/fae/HMG.htm
[email protected]
-----Original Message-----
From: morphmet <[email protected]>
To: morphmet <[email protected]>
Sent: Mon, Apr 9, 2012 1:56 pm
Subject: Next Engine scanner protocol
-------- Original Message --------
Subject: Next Engine scanner protocol
Date: Mon, 9 Apr 2012 18:01:31 +0200
From: Javier Santos<[email protected]
<mailto:[email protected]>>
To: Morphomet New Mailing List<[email protected]
<mailto:[email protected]>>
Hello morphometricians,
I am working with snow vole skulls and have been provided a Next Engine
surface scanner to obtain my shape data. I have done various tests with
the scanner, but I still don't achieve the resolution I was expecting in
the quality of the scans after alignment. I was hoping that somebody who
has used or is using this scanner could give me some tips on how they
program the scanner to obtain their images, such as number of divisions,
starting angle, starting tilt, etc. As well, any tip on what to use to
position and maintain the skull still during scanning (especially as I
use the optional robotic arm for the Next Engine scanner). I would also
appreciate any other useful information you might know on how to obtain
fine morphological data from micromammal skulls using other techniques,
software, etc. Thank you a lot in advance!
All the best,
Javier Santos
Museo Nacional de Ciencias Naturales (MNCN-CSIC)
Biodiversity and Evolutionary Biology Department
