Thanks for that, Even & Andre, There is a NITF_ABPP metadata item and that is set to 11 so that explains why all leading bits are 0. Also min/max on the stats is 1 to 2047.
The reason why I was going on about 5 bits for R etc was because that is how colours are defined in a 16 bit bitmap and assumed that was how the data was going to be organised in a 16 bit single band (if it was gray-scale, I assumed that the ColorInterp would be set to gray-scale based on the NITF_IREP=MONO metadata item). I guess I'll have to grab the histogram and scale the pixel values accordingly into an 8 bit gray-scale........I may be back!! Thanks Martin -- View this message in context: http://osgeo-org.1560.x6.nabble.com/16bit-Pixel-Arrangement-tp5166812p5167148.html Sent from the GDAL - Dev mailing list archive at Nabble.com. _______________________________________________ gdal-dev mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/gdal-dev
