Orfeo ToolBox (Open Source processing of remote sensing images)
Orfeo ToolBox or OTB is an open source application for processing high resolution optical, multispectral and radar images at the terabyte scale.
Orfeo ToolBox is available in the following versions:
Loading Orfeo ToolBox
Orfeo ToolBox can be loaded with
module load otb
Using the command line tools
You can find the numerous OTB applications and examples of them from the OTB CookBook
For example, you can calculate NDVI from a Sentinel virtual raster by using the otbcli_RadiometricIndices application. This requires NIR band to be in the first channel and RED band in the second
otbcli_RadiometricIndices -in <virtual raster> -channels.nir 1 -channels.red 2 -list Vegetation:NDVI -out <output_file>
Using the OTB Python tools
The applications included in OTB can also be run by using the Python bindings included in the module.
Here's an example how to run mean raster calculation on a test virtual raster with Python.
import otbApplication as otb fp = <INPUT RASTER FILEPATH> out = <OUTPUT RASTER FILEPATH> app = otb.Registry.CreateApplication("Smoothing") app.SetParameterString("in", fp) app.SetParameterString("type", "mean") app.SetParameterString("out", out) app.ExecuteAndWriteOutput()
If you need additional libraries installed to the Python environment, contact firstname.lastname@example.org
Running OTB applications in parallel
OTB applications seem to scale the number of processing threads automatically which means that the applications generally run faster when given more CPU cores.
Here is an example batch job script with 4 CPU cores
#!/bin/bash #SBATCH --job-name=<name_of_your_job> #SBATCH --account=<your_project> #SBATCH --time=00:03:00 #SBATCH --ntasks=1 #SBATCH --cpus-per-task=4 #SBATCH --mem-per-cpu=2000 #SBATCH --partition=small module load otb otbcli_RadiometricIndices -in test_vrt.vrt -channels.nir 1 -channels.red 2 -list Vegetation:NDVI -out NDVI.tif
It is also possible to run OTB in parallel in several computing nodes using MPI but it has not been tested yet on Puhti
License and citing
Orfeo ToolBox is licensed under the Apache License, Version 2.0. The full licence can be found here
In your publications please acknowledge also oGIIR and CSC, for example “The authors wish to acknowledge for computational resources CSC – IT Center for Science, Finland (urn:nbn:fi:research-infras-2016072531) and the Open Geospatial Information Infrastructure for Research (oGIIR, urn:nbn:fi:research-infras-2016072513).”
Last edited Mon Dec 30 2019