[gdal-dev] Problem fixing the Starting point of a tiling job

[Moses.Gone at t-systems.com]

Dear Experts,

When tiling datasets, we want to specifically define custom windows. To do this, we are using -te option during warping. The warped output is then tiled.

The problem though is that after tiling and checking the bbox of the first tile, we have noticed that the starting coordinate is shifted and do not correspond to the defined starting coordinates as illustrated below:


*       The blue grid represent the original dataset with cellsize (a).
*       The black grid represent the new dataset with cellsize (b).
*       The cellsize of both datasets can be the same or not [(a) .eq. (b) or (a) .ne. (b)].
*       The red polygon represent the bbox which fit to the new cellsize and tilesize.

We would like to start tiling exactly at the upper left corner of the red polygon.How we can force gdal to clip the raster exactly according to our well defined bbox?

In some cases, GDAL picks the right Bbox startpoint, in other cases it doesnt. Is this somehow related to the difference in the input and output data resolution(cellsize)?

We can't figure out why it is ignoring our definitions. Could someone help shade some light to what is happening here?

Thanks in advance,
Moses

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20140312/6f73b31c/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BBox_def.png
Type: image/png
Size: 9698 bytes
Desc: BBox_def.png
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20140312/6f73b31c/attachment-0001.png>