[TriLUG] this is a serious question - how to detect things in photos with Linux

Charles Fischer cfischer at modernferrotype.com
Sat Dec 8 12:19:57 EST 2012

If you use OpenCV as the recognizer, you may want to use the C++
interface.  IIRC it is more complete than the Perl interface.  You will
have one big challenge using OpenCV.  It needs a lot of images for
training.  On the order of 5 to 10 thousand of both positive and negative
images.  Not to hard to find a face data base, but finding enough naughty
parts images with the correct copyright could be a problem.  It is also
very time consuming to have a UP mark up all of the images.

Thanks for the UPF acronym.

Good luck,
Charles Fischer

On Sat, Dec 8, 2012 at 11:00 AM, Peter Neilson <neilson at windstream.net>wrote:

> On Sat, 08 Dec 2012 09:30:22 -0500, Matt Flyer <matt at noway2.thruhere.net>
> wrote:
>  I think this part is a good idea.  Face recognition algorithms ...
>> ... Maybe even consider 'no face' = 'no sanppy'?
> Junknose face okay? I can think of four ways to do that without even
> drinking.
> Oh, I forgot, the UPF will be the second stage of this device.
> --
> This message was sent to: Charles Fischer <cfischer at modernferrotype.com>
> To unsubscribe, send a blank message to trilug-leave at trilug.org from that
> address.
> TriLUG mailing list : http://www.trilug.org/mailman/**listinfo/trilug<http://www.trilug.org/mailman/listinfo/trilug>
> Unsubscribe or edit options on the web  : http://www.trilug.org/mailman/**
> options/trilug/cfischer%**40modernferrotype.com<http://www.trilug.org/mailman/options/trilug/cfischer%40modernferrotype.com>
> TriLUG FAQ          : http://www.trilug.org/wiki/**
> Frequently_Asked_Questions<http://www.trilug.org/wiki/Frequently_Asked_Questions>

More information about the TriLUG mailing list