<p>Richard, I can strip the crs off using gdal_translate and setting a Baseline profile. But I wanted to test what happens when a raster is already georeferenced.</p>
<p>Giovanni</p>
<p>Inviato da dispositivo mobile</p>
<div class="gmail_quote">Il giorno 26/ott/2011 09.23, "Richard Duivenvoorde" <<a href="mailto:rdmailings@duif.net">rdmailings@duif.net</a>> ha scritto:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On 2011-10-26 00:42, G. Allegri wrote:<br>
> Ok, i got it.<br>
> The reason it doesn't work is probably easily explained: the source X/Y<br>
> are considered as raster space coordinates, while I thought that having<br>
> the raster a CRS assigned this was kept into account during the warping.<br>
> Probably manual re-georeferencing is conceptually unuseful, outside my<br>
> exercise. Even if reprojection wasn't enough, the best way would be to<br>
> remove any CRS reference from it.<br>
> Tomorrow I will do more tests, but I'm pretty sure that I was simply<br>
> trying a wrong procedure...<br>
<br>
I think qgis/ogr takes your crs into account. But if you open your<br>
raster (tiff) in Gimp, and save it again as a tiff, Gimp will 'help' you<br>
to remove the crs information from your geo(tiff), giving you a plane<br>
raster with not crs information to do your tests :-)<br>
<br>
R<br>
</blockquote></div>