[GRASS-dev] i.points, i.vpoints now use G_parser()

[Glynn Clements]

Michael Barton wrote:

> So I should display a forward RMS and a backward RMS.
> 
> For each point, should I just display a single error distance or try to
> display reverse for raster and forward for vector?

I suspect that you should probably display both.

AFAICT, either of the two points in a GCP pair can be misplaced just
as easily. Also, when you are entering the control points, do you
actually know whether they are going to be used for transforming a
raster or a vector (or both)?

Maybe you could determine the size of the source and destination
regions, and convert the absolute errors to a proportion of the
region's size. That would allow you to sensibly compare forward and
backward errors.

-- 
Glynn Clements <glynn at gclements.plus.com>