Hi Michele, On Fri, Jul 29, 2016 at 6:05 PM, Michele Toma <[email protected]> wrote: > Hi Moritz/everyone, > > I really appreciate all of the information. > > The original issue was exporting an RGB composite image from GRASS to a > GeoTIFF format that could be opened in Global Mapper. I had created a > pansharpened RGB image and attempted to export the image, but the settings I > used created entirely black images.
It is important to understand if there was really no data or simply a wrong color table applied. > When I exported using r.out.gdal with no create options, the resulting image > would not open in Global Mapper. I then tried to export without the color > table which gave me a grayscale image. So, my understanding was that the data > type was wrong. Unfortunately - in addition - GeoTIFF cannot represent all data types. > The errors I got when trying to export as Byte said to try using Float64 You could round the values after having multiplied them by 10 or 100. I agree, too many possibilities existing... >. But after trying to use that data type and all other data types in the drop >down list, I ended up with a black image. In addition to changing the data >type, I opted to skip exporting the color table according to the directions on >the r.out.gdal documentation page. What options should I be selecting to >export the image? Is there a simpler method to exporting the image that I am >not aware of? Maybe other users have a suggestion here. > When trying to export in QGIS, I use GRASS tools to open the rbg composite > raster and then right click on the layer -> Save As... The options that I use > are "export as a rendered image" and in create options: "No compression". > This seems to work for me. Fine. > However, I have run into another issue where I cannot get rid of the black > borders/null data that surround the image. The borders are removed in GRASS > after running r.null on the rgb composite image. They reappear when opening > in QGIS using GRASS tools. (Note: I am using QGIS 2.8.2). For this here a method, it uses r.clump: http://courses.neteler.org/selective-data-removal-elevation-map-floodfilling/ (in my example for elevation models but it works the same for imagery). Markus _______________________________________________ grass-user mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-user
