Re: [RFC] New task: science-dataacquisition
On Tue, 9 Dec 2008, Gerber van der Graaf wrote:
A whole area of science that depend on image analyses (from microscopic
to astronomics) might like to include different programs to control
cameras. Some ideas:
In general I like the idea to give the whole bunch of image acquistion
programs some structure and give some guidelines which might be relevant
for scientific purposes. But IMHO this would stretch the scope of
dataacquisition to large and I would prefer a separate imageacqusition
task. Well, I know images are data as well, but I think a split between
numeric data and image data makes perfectly sense from a user point of
view.
So I'm open for suggestions if anybody wants to provide reasonable
dependencies for a potential imageacqusition task.
Firewire/IEEE1394 kernel modules, libraries and the coriander program.
Though I have the impression that this interface is currently under
transition and not working properly with the Linux kernel currently
included in Debian/Unstable:
http://ieee1394.wiki.kernel.org/index.php/Juju_Migration
Anyhow the Firewire interface (and coriander) support many cameras.
Video4Linux: an apt-get search gave me a whole bunch of libraries and
applications, partly overlapping Firewire. Maybe only a part of it will
be sufficient.
Yes. We should probably put some barrier here to suggest only those
packages that are really useful for scientific research.
Probably there is more for acquiring images from cameras available on
GNU/Debian, but I think this will cover the most important ones.
That's the whole point of the Blends effort: Detect and suggest reasonable
packages out of the flat and unstructured pool for a certain user group.
Kind regards
Andreas.
--
http://fam-tille.de
Reply to: