<div>Hi,</div><div> </div><div>I have a huge tiff image (covering an entire country) with 3399 columns and 3747 rows, so a total of 12,736,053 cells. The cell size is 1000m or 1km. I need to extract the data per km2. Of course, I didn't think it was smart to work with such a huge dataset in one go. So, I followd these steps in Quantum gis -</div>
<div> </div><div>1. I used the<strong> gdalwarp -crop_to_cutline</strong> in the command line to carve out a portion of the raster image based on the shapefile of a state of the entire country. <strong>The problem is that although the gdalwarp command worked it is clearly causing a shift (of about 500m) during this step.</strong> </div>
<div> </div><div>2. Then, I used the <strong>gdal2xyz tool</strong> from the command line to create the csv file corresponding to the 'cut out' raster image or only the raster image of the state.</div><div> </div>
<div>3. Then I used the 'Create a Layer from a Delimited Text file' plugin in qgis to get the centroid shapefile. </div><div> </div><div>My problems are-</div><div> </div><div>1. <strong>The shift in the image caused by the gdalward tool. Does anyone know how to work around that, so that no shift is caused?</strong></div>
<div> </div><div>2. The image is in 32 bit floating point and so does not show the attribute table. Even after getting the csv file I am unable to link the created csv file with the raster image, and thus cannot understand which pixel corresponds to which value in the csv file. </div>
<div> </div><div>Can anyone suggest any better way of doing this? </div><div> </div><div>Thanks very much for your help.</div><div> </div><div>Best Regards,</div><div> </div><div>Tilo</div><div><br clear="all"><br>-- <br>
</div>