University Project (FSU Jena) by Patrick Fischer, Marlin M. Mueller and Jonas Ziemer
This tool combines pyroSAR Sentinel-1 preprocessing capabilites with GRASS GIS (Version: 7.9dev) functionality for Sentinel satellite imagery.
It allows for automatic downloading of user-defined Sentinel-1 scenes, pre-processing, subsetting and multitemporal operations. Data is handled in GRASS internal formats and all available GRASS operations can be applied
- Automatic Setup of GRASS environment
- Automatic download of Sentinel-1 data using the GRASS addon i.sentinel.download
- Pre-processing of downloaded Sentinel-1 using pyroSAR functionality (ESA SNAP is required for the processing, see here)
- Automatic subsetting of satellite data to specified region of interest
- GRASS functionality:
- Creation of space time raster datasets (STRDS) in GRASS
- Different multitemporal operations in GRASS
- Future functionality:
- pyWPS implementation
Developed in Python 3.6, currently only supported on linux-based OS
In case you have git installed you can install the package as follows:
pip install git+https://github.com/marlinmm/GRASSINEL.git
GRASSINEL workflow
Example of created timeseries processed using pyroSAR and created using GRASS GIS time series functionality
Sentinel-1 time series from 2020-06-06 to 2020-07-29 using only relative orbit number 168 in desending direction with
20m resolution with terrain flattening and thermal noise removal active (roughly 5 mins on 14 cores per scene)
Sentinel-1 time series from 2020-06-01 to 2020-06-07 using all available datasets for the chosen time frame in
ascending direction with 50m resolution without terrain flattening and thermal noise removal (roughly 2 mins on 8 cores
per scene)
Point plot of Sen-1 acquisition dates for VV and VH polariazation printed by the "visualize_stc" function
Screenshot of the GRASS addon for pyroSAR processing inside GRASS GISS