[Qgis-user] Georeferencing Problem in QGIS-Version 3.34.12-Prizren

[Greg Troxel]

Belal Elsharkawy via QGIS-User <qgis-user at lists.osgeo.org> writes:

> if anyone has an idea or can help, I have a problem with georeferencing any 2D plan since a long time and have tried almost everything possible.
> After importing the georeferencing wanted plan in the Georeferencing
> window in QGIS and choosing 3 points and their representatives in the
> main window plan, then going to the settings in the Georeferencing
> window to check the coordinate system and make sure that it’s the same
> as the one used in the main QGIS window as well as the Location where
> it will be saved.
> I always get a very big differences in the dx (Pixel) as well as the dy (Pixel) which they should be zero or near to zero  (see the screenshot attached) and it doesn’t work.
> While other colleagues doing literally the same as I do, and it works for them.

I wonder if you somehow have an option checked not to allow rotation.

I would suggest putting in more points, at least 5 maybe even 10, and
then look at patterns in the residuals.

What CRS is the project in?  If it were in a geographic CRS instead of a
projected (and conformal) CRS, I could see that causing problems.
Personally I tend to use a UTM zone of my national datum, or a lambert
conformal conic ("state plane").  I would expect you have some similar
CRS which fits local norms, perhaps UTM on ETRS.