Command line image processing tools, developped for the first version of the Co-ReSyf platform.
coresyf_toolkit
├── src -> contains the scripts of all tools
└── examples -> folder where the example data files are stored
CORESYF TOOLKIT works with Python 2.6-2.7
No need to install anything. Just run docker (see usage for examples).
If the image is not yet built, run the command:
docker-compose build.
This will create an image with the tag repo.coresyf.eu/toolkit:latest.
This will find all tests.py scripts with unittest module and test the tools:
docker-compose run test.
To test docker-compose up <tool_name>. Tool name can be one of the following:
- calibration: coresyf_calibration
- crop: coresyf_image_crop
- mask: coresyf_image_mask
- spliting: coresyf_image_splitting
- stacking: coresyf_image_stacking
- isodata: coresyf_isodata_classification
- pt2grid: coresyf_pointsToGrid
- polygon2polyline: coresyf_PolygonToPolyline
- polyline2raster: coresyf_PolylineToRaster
- randraster: coresyf_randRasterGen
- raster2polygon: coresyf_RasterToPolygon
- sarfile: coresyf_sar_fileSelector
- speckle: coresyf_speckle_fileter
- vector: coresyf_vector_creator
NOTE: Image crop is the only tool implemented, and it is not working (needs solving).
It's possible to test the tools without running any of the predefined commands and options. To do so, invoke the tool with the following command:
docker container run repo.coresyf.eu/toolkit <tool_script>.py
The Dockerfile/docker-compose files are configured to place scripts and examples inside the docker container in the following directories:
- src:
/opt/toolkit/src - examples:
/opt/toolkit/examples
Note that, while source code is mounted as a volume and copied inside the container, the examples folder is solely mounted as a volume.
Also, be aware that the default woking dir of the container is /opt/toolkit/src, so there is no need to specify a path to the python script.
To add a new python package requirement, simply add it to the requirements.txt file and rebuild the image.
These tools (coming from the requirements assessed for the different Research Applications) will be deployable in the cloud back-end through either calls from the Workflow Manager or the Developer’s Sandbox.
-
Radiometric Correction: an image pre-processing module to perform radiometric correction of both optical and SAR imagery.
-
Optimal Interpolation: an image processing module to support the implementation of optimal interpolation methods to obtain gridded maps from point measurements.
-
Image crop: an image processing module to support the possibility of cropping an image for a specific AOI.
-
Image mask: an image processing module to support the masking of raster imagery using either other raster imagery r vector files (e.g. , shapefiles)
-
SAR Speckle filtering: an image pre-processing module for SAR imagery to remove the speckle effect.
-
Image statistics: an image processing module to support the calculation of general image values statistics.
-
Geo-referencing: an image processing module to support the geo-referencing of optical and SAR satellite imagery
-
Error metrics: an image processing module to support the calculation of different error metrics
-
Vector creation and edition: an image processing module to support the creation and edition of different vector file formats
-
Layer stack creation: an image processing module to support the ability to stack images into one multi-layer file sorted in the order in which they are inserted by the user or by any other user specific parameter (e.g. time)
TBD
TBD
TBD
TBD
TBD
TBD
TBD
TBD
TBD
TBD
- V0.2: December 14, 2017: Dockerization
- Dockerized toolkit
- Improved readme
- V0.1: January 16, 2017: first release:
- ADDED: Image Crop, Image Mask, Image Splitting, Image Stacking and Points-to-Grid tools.