[GRASS-dev] Re: [GRASS-PSC] i.ortho.photo

[Brad Douglas]

On Thu, 2006-12-14 at 21:48 -0700, Michael Barton wrote:
> Brad,
> 
> I haven't used the module, so I can't say. The reason for making it
> *possible* to run non-interactively (even if it *can* be run interactively
> if desired) is that it can then be wrapped into a non-x11 GUI. As long as
> interactive terminal input is required, this is not possible.

Excellent point.  I hadn't considered alternative GUIs.

> There is no problem running the module twice in different modes to have it
> compute transform coordinates, so long as those coordinates can be output to
> stdout or a file that could be subsequently used by the module running in
> different mode. On the other hand, if this involves interactively clicking
> on photo reference points and a georeferenced map to create a file of point
> pairs (photo/image and georeferenced equivalents), there should be the
> option of replacing this task by GUI code. For example, the TclTk
> georectifier integrates i.group, i.target, and i.rectify. However, the
> interactive i.points is simply replaced by TclTk code that generates a POINT
> file used by i.rectify.

Right now, coordinates are stored in the REF file.  I assume the same is
true for i.points/i.rectify, but haven't reviewed that code, yet.

> I'm not sure if this is helpful to you or not. However, to incorporate this
> into a current or new GUI, there needs to be some way to run this module or
> its decendent(s) 100% non-interactively.

Yes, this has been helpful.

> Michael
> 
> 
> On 12/14/06 9:22 PM, "Brad Douglas" <rez at touchofmadness.com> wrote:
> 
> > I've thought about adding options for input transforms coordinates, but
> > I'm not certain how useful it would be.  Do you think it's something
> > that would be useful?  How often would users know ahead of time what the
> > coordinates were?  Perhaps for re-projection?
> > 
> > One of the goals is to remove as much interactivity as possible, so I'm
> > quite interested in alternatives.  I still intend on leaving that
> > portion interactive, but have considering "bypass" options.
> > 
> > On Thu, 2006-12-14 at 21:11 -0700, Michael Barton wrote:
> >> My only comment is to encourage you in trying to find a way to make
> >> i.orthophoto work in a non-interactive mode, at least as one option. As to
> >> the C code changes, I cannot comment.
> >> 
> >> Michael
> >> 
> >> 
> >> On 12/14/06 7:59 PM, "Brad Douglas" <rez at touchofmadness.com> wrote:
> >> 
> >>> 
> >>>     dem_opt->required      = NO;
> >>>     dem_opt->multiple      = NO;
> >>>     dem_opt->description   = _("Digital Elevation Model");
> >> 
> >> __________________________________________
> >> Michael Barton, Professor of Anthropology
> >> School of Human Evolution & Social Change
> >> Center for Social Dynamics & Complexity
> >> Arizona State University
> >> 
> >> phone: 480-965-6213
> >> fax: 480-965-7671
> >> www: http://www.public.asu.edu/~cmbarton


-- 
Brad Douglas <rez touchofmadness com>                    KB8UYR/6
Address: 37.493,-121.924 / WGS84    National Map Corps #TNMC-3785