[gdal-dev] '-mask' option of gdal_translate doesn't work
Hello, I tried command line below that is copied from http://www.gdal.org/gdal_translate.html: gdal_translate rgba.tif withmask.tif -b 1 -b 2 -b 3 -mask 4 -co COMPRESS=JPEG -co PHOTOMETRIC=YCBCR --config GDAL_TIFF_INTERNAL_MASK YES It doesn’t generate transparency channel in output. But if I removed options “-b 1 –b 2 –b 3”, the output image ‘withmask.tif’ would contain transparency channel, that could be observed by opening the image in the Photoshop. But if I remove all options to modify command line like this: gdal_translate rgba.tif withmask.tif The transparency channel is still generated in output even the option ‘-mask’ is not specified at all. And the last alpha channel in the source image is always picked up as transparency channel to the output no matter what band number is given to the ‘-mask’ option. Then I tried the same command line on a new RGB tiff that has more than 4 channels and output image doesn’t contains any transparency channel even ‘-mask band’ option is explicitly specified. It seems that the option ‘-mask’ doesn’t work and generating the transparency channel is only depends on the colorspace (RGBA or Greyscale + alpha). Please let me know how to correctly use this option. Thanks! Peter Z___ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
Re: [gdal-dev] How to cut a Raster map with a KML file
Luis, On Tue, Sep 11, 2012 at 11:14 AM, Luis Lisboa luislisboa1...@gmail.com wrote: Hi Eli Probably I was not clear: When I use this it outputs a geotiff with two bands: 1 with the real data and another one with the mask of the pixels that fall inside each polygon. I just want to have an output file with one band (band 1) and without the mask. Is it possible? Sometime images can clean up just be running them through gdal_translate. -ot or -b or -mask or some combination may additionally be needed. http://gdal.org/gdal_translate.html Does gdal_translate on your file collapse the mask band into a single band with no data values? rgb2pct.py can be used to convert RGB/RGBA to single band with a color table. I'm not sure if it would work with a two band image, one being the mask but you can try it and see what it does. HTH, Eli Thanks Luis On Thu, Sep 6, 2012 at 4:50 AM, Eli Adam ea...@co.lincoln.or.us wrote: Luis, On Wed, Sep 5, 2012 at 3:00 AM, Luis Lisboa luislisboa1...@gmail.com wrote: Thank you Chitanya. But it created an output file with two layers (one with the values from original file and another with a 255 for the pixels thar are inside KMZ file: How can I have produce one layer? I may be thinking of something else based on your description. If you got an output that has the original values in the area of the kmz and an alpha mask band over the other areas I think that would be expected. You can also use -crop_to_cutline to remove that 'extra' data. If that isn't what you are talking about, you may want to use ogr2ogr to convert your kmz to a shapefile in the same projection as your input raster and see if it gives you a better result. Bests, Eli Thanks Luis On Mon, Sep 3, 2012 at 5:10 PM, Chaitanya kumar CH chaitanya...@gmail.com wrote: Luis, gdalwarp has some options for this exact purpose. The cutline can be from any of the OGR readable datasource. http://www.gdal.org/gdalwarp.html On Mon, Sep 3, 2012 at 9:34 PM, Luis Lisboa luislisboa1...@gmail.com wrote: Greetings I have a raster file (Geotiff) and I have a KML file with a polygon and I want to cut raster map to only have valid values inside the polygon. How can I do this using just GDAL? Usually I used to cut geotiff file with gdalwarp and but in those cases I was cutting a square (xmin xmax ymin ymax) and in this case is a more complex polygon Thanks Regards, Luis ___ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev -- Best regards, Chaitanya kumar CH. +91-9494447584 17.2416N 80.1426E ___ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev ___ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
[gdal-dev] MUTEX_NONE question
If I don't want to use GDAL in thread safe mode on Windows can just define MUTEX_NONE in the nmake.opt file? If not, is there a way I can do it. Also, I want to avoid caching. Can I just define the env var GDAL_FORCE_CACHING and set it to NO? Will that also avoid any thread synchronization? Best regards, Martin Chapman Software Developer Pixia Corp. 45615 Willow Pond Plaza Sterling, VA 20164 Main - 571.203.9665 Cell - 303.324.1065 chapm...@pixia.com The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. ___ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
Re: [gdal-dev] MUTEX_NONE question
On Tue, Sep 11, 2012 at 1:08 PM, Martin Chapman chapm...@pixia.com wrote: If I don’t want to use GDAL in thread safe mode on Windows can just define MUTEX_NONE in the nmake.opt file? If not, is there a way I can do it. Martin, I believe so, yes. Also, I want to avoid caching. Can I just define the env var GDAL_FORCE_CACHING and set it to NO? No, I don't believe that will help much. This configuration options just exists (I think) to force use of caching for drivers or situations where the native driver would normally avoid it. I am not aware of a build or configuration option you can set to avoid caching. If you really really want to avoid caching you can try to directly call GDALRasterBand::ReadBlock() which will completely avoid the block cache. I haven't done anything like that for a long time so it may cause problems for some drivers. Will that also avoid any thread synchronization? The MUTEX_NONE should stop essentially all thread synchronization. But if you are using multiple threads you may encounter serious bugs with mutexes disabled. Doing this makes the most case in single threaded apps that want to avoid the operating system mutex overhead. Best regards, -- ---+-- I set the clouds in motion - turn up | Frank Warmerdam, warmer...@pobox.com light and sound - activate the windows | http://pobox.com/~warmerdam and watch the world go round - Rush| Geospatial Software Developer ___ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
