Le jeudi 03 septembre 2015 09:03:29, Andrea Peri a écrit : > Hi, > I have a VRT datasource, with more raster inside. > Each raster is partially overlapped to the two near rasters. > > My dubt is how the histogram retrieved using the API GetDefaultHistogram, > is counting the overlapping. > > Infact the overlapping portin is more NULL values, but also there is > portion f informations. > > Is in a pixel overlapped there is a NULL from frame1 and a VALUE from > frame2. I could guess the Histogram ignore the NULL from frame1 and count > only the VALUE from frame2. > > But when the frame1 and frame2 give the differents values: VALUE1 and > VALUE2. In this case what is the value counted in the histogram ? > > It count the top level raster only, the first raster only, use a mean > of both or count both duplicating so the number of cell in the results > ?
Andrea, Hopefully (and that's what the existing code is supposed to do), the histogram reported on the VRT should be the same as the one you would get on a GeoTIFF that would result from the conversion of the VRT. Said other wise, it should be the histogram of what is displayed. Even -- Spatialys - Geospatial professional services http://www.spatialys.com _______________________________________________ gdal-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/gdal-dev
