[Fwd: Re: [ptx] Stitching scanned, "flat" images]

Emmanuel emmanuel at favre-nicolin.net
Sun Oct 10 04:07:49 BST 2004


I don't know if it is parctical but there is a project on sourceforge 
dedicated to what you're looking at : xmerge


Le mardi 05 Octobre 2004 12:44, HaJo Schatz a écrit :
> I could slap my forehead for that... Of course you're right, tnx a lot!
>
> I do have some difficulties with Hugin doing this however, as it doesn't
> seem to allow such things as optimizing d/e for each individual image.
> But with a manually created optimizer script things work amazingly well.
>
> On Tue, 2004-10-05 at 21:30, Jim Watters wrote:
> > The proper way to do this with PT is to use the offset variables keeping
> > y,p,r constant just optimize d & e
> >
> > # d,e          initial lens offset in pixels(defaults d0 e0, optional).
> > #                  Used to correct for offset from center of image
> > #                  d - horizontal offset,
> > #                  e - vertical offset
> >
> >
> >  Jim Watters
> >
> >  Graphic Software Developer
> >  http://photocreations.ca
> >
> > HaJo Schatz wrote:
> > >Hi List,
> > >
> > >Maybe a silly question but -- could Hugin also be used to stitch
> > >scanned, "flat" images?
> > >
> > >What I'm trying to achieve is to scan a large map in multi-row pieces
> > >and stitch it together. There'll be yaw/pitch/roll corrections necessary
> > >but nothing else I assume. The projection is different from a panorama I
> > >guess, as the scale has to remain correct on all parts of the map after
> > >scanning (I'll be using the result for navigational purposes, Ie I can
> > >not affort distortions being introduced).
> > >
> > >Any thoughts or pointers to alternative methods are highly welcome. I
> > >have tried xmerge but I'm not really happy with the results, it seems to
> > >introduce errors of a few pixels at least.
> > >
> > >Tnx,
> > >HaJo


More information about the ptX mailing list