Forum

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - calculon

Pages: [1]
1
Bug Reports / Re: Export fails in cluster if path contains variables
« on: April 13, 2023, 12:52:24 PM »
In case anyone is facing a similar problem:

We recently managed to fix it by using <path>/{projectfolder}/dem_{chunklabel}.tif</path> in our batch file.
The first slash seemed to make the difference for us.

2
Bug Reports / Export fails in cluster if path contains variables
« on: March 09, 2022, 05:07:29 PM »
We run batch processes in cluster and found, that TIF exports of DEM and ORTHO does not work.
The reason seems to be variables in the output paths, absolute paths work as expected.
There is no error messages or other indications of fail, the process simply does not yield expected results.

Here is the diff between our non-workin and working versions (absolute paths work via cloud, the others not):

diff ms_batch.xml tmp.debug_ms-tiffexportcloud.xml
61c61
<     <path>{projectfolder}/dem_{chunklabel}.tif</path>
---
>     <path>/data/tmp-data/dem_debugout.tif</path>
81c81
<     <path>{projectfolder}/ortho_{chunklabel}.tif</path>
---
>     <path>/data/tmp-data/ortho_debugout.tif</path>

3
Reading this thread again led me to the solution, thank you for fixing this in Version 1.6.5, just tried it.

Quote
Quote
Of course there are many other options neglected in .xml for Export Orthomosaic settings e.g. JPEG compression value, Generate Tiff Overview, Write alpha channel, etc.
The reported issues with Compression, Region and no-data sections should be fixed in the 1.6.5 version update.

We were missing the BigTIFF parameter in the batch xml. It does not get saved in any way and is not read in again.
I correctly guessed it should be something like tiff_big, and finally these parameters appeared:
Code: [Select]
    <image_compression>
       <tiff_big>false</tiff_big>
       <tiff_tiled>false</tiff_tiled>
       <tiff_overviews>false</tiff_overviews>
       <tiff_compression>5</tiff_compression>
       <jpeg_quality>90</jpeg_quality>
     </image_compression>
None of the image_compression options seemed ever to get saved or read in before v. 1.6.5.
Now they do!

To use it in a python based workflow see https://www.agisoft.com/forum/index.php?topic=11687.msg52407#msg52407

Code: [Select]
compression = Metashape.ImageCompression()
compression.tiff_big = True
chunk.exportRaster(path = "path/to/file.tif", source_data=Metashape.OrthomosaicData, image_compression = compression)

Pages: [1]