<div dir="ltr"><div class="gmail_extra"><div class="gmail_default" style="font-family:tahoma,sans-serif">Adrien,<br><br></div><div class="gmail_default" style="font-family:tahoma,sans-serif">OGC in its GML standard document -- of course in line with ISO -- recommends that:<em> <br><br>"When a grid point is used to represent a sample space (e.g. image
pixel), the grid point represents the center of the sample space (see
ISO 19123:2005, 8.2.2)"</em>.
</div><br><div class="gmail_default" style="font-family:tahoma,sans-serif">GML encodes the locations of a raster/gridded-coverage by means of a grid of 0D points.<br></div><br><div class="gmail_default" style="font-family:tahoma,sans-serif">You can always put your raster origin wherever you want, i.e. in whatever corner you prefer. <br>Alas, there's no way -- that I am aware of -- to specify what is the shape of a measurement footprint/sample-space, nor to say if the footprint surrounds the point-geometry equally (point is pixel-centre) or some other way (point is pixel-upper-left-corner?).<br><br></div><div class="gmail_default" style="font-family:tahoma,sans-serif">Hence raster creators should probably stick to this recommended encoding until standards do not offer more freedom on this :)<br></div><br><div class="gmail_default" style="font-family:tahoma,sans-serif">@seeAlso<br></div><div class="gmail_default" style="font-family:tahoma,sans-serif"><a href="http://rasdaman.org/wiki/PetascopeSubsets">http://rasdaman.org/wiki/PetascopeSubsets</a><br><a href="http://rasdaman.org/ticket/680">http://rasdaman.org/ticket/680</a></div><br>-<div class="gmail_default" style="font-family:tahoma,sans-serif;display:inline">Piero !</div><br><br><div class="gmail_quote">On 29 June 2015 at 21:00, <span dir="ltr"><<a href="mailto:gdal-dev-request@lists.osgeo.org" target="_blank">gdal-dev-request@lists.osgeo.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div id=":20m" class="" style="overflow:hidden">Date: Mon, 29 Jun 2015 11:12:11 +0200<br>
From: Adrien ANDR? <<a href="mailto:adr.andre@laposte.net">adr.andre@laposte.net</a>><br>
To: <a href="mailto:gdal-dev@lists.osgeo.org">gdal-dev@lists.osgeo.org</a><br>
Subject: [gdal-dev] Raster origin coodinates in file header<br>
Message-ID: <<a href="mailto:55910BEB.2030307@laposte.net">55910BEB.2030307@laposte.net</a>><br>
Content-Type: text/plain; charset=utf-8; format=flowed<br>
<br>
Dear list,<br>
<br>
opening 2 rasters with different resolutions i noticed a shift.<br>
<br>
Asking our imagery provider, i was told the origin coordinates they<br>
store in image header (JP2 files) are those of the top left pixel center.<br>
<br>
GDAL reads origin coordinates as the top left pixel top left corner<br>
coordinates, as i'm used to.<br>
<br>
<br>
Of course, i fixed my problem, the hacky way, playing with<br>
gdal_translate a_ullr option, but:<br>
<br>
How to deal with this?<br>
How not to worry about georeferencing each time i load a raster file?<br>
Is there any OGC recommendation about origin coordinates storage?<br>
<br>
<br>
Thanking you in advance,<br>
<br>
Regards,<br>
<br>
Adrien</div></blockquote></div><br><br></div></div>