Skip to content

open-science-data/alignSAR

 
 

Repository files navigation

AlignSAR

The AlignSAR tools are used to extract representative SAR signatures, and ultimately offer FAIR-guided open SAR benchmark dataset library designed for SAR-based artificial intelligence applications, while ensuring interoperability and consistency with existing and upcoming initiatives and technologies, facilitating wider exploitation of SAR data and its integration and combination with other datasets. This library will contain meaningful and accurate SAR signatures created by integrating and aligning multi-SAR images and other geodetic measurements in time and space. Related link: https://www.alignsar.nl

Tool description:

  1. MLscripts: scripts for machine learning analysis. Yolov8, ANN and Siamese is seperately used for Object Detection (India), Land Use Land Cover classification (Netherlands), and Change Detection (Polands).
  2. bin: non-python scripts used within the toolbox (e.g. rdr<->geocode)
  3. itc-doris_5_patch2023: an updated version of Doris-5 software developed by TUDelft.
  4. jupyter_notebook_demo: a jupyter notebook to demonstrate how to extract, visualize and analyse SAR signatures.
  5. rdcode: the radarcoding scripts.
  6. snap_graphs: graphs for SNAP used within the toolbox (e.g. rdr<->geocode)
  7. stac: python scripts to create STAC
  8. AlignSAR_tutorial.pdf: tutorial for this package
  9. DockerFile: docker file with OS and software setup. Note that another docker file with Doris-5 installed is in itc-doris_5_patch2023, with the same name DockerFile.
  10. Meta_info_extraction_global_local.py: script to extract global/local attributes from Sentinel-1 SAR metadata.
  11. bashrc_alignsar: install settings for the expected environment variables and paths
  12. resdata.py: script needed from signature_extraction.py.
  13. signature_extraction.py: SAR signature extraction script.
  14. speckle_filt.py: A speckle filtering script.
  15. alignsar_utils.py: various python functions used within the toolbox.

Note that this is research code provided to you "as is" with no warranties of correctness. Use at your own risk.

Installation

Either build a docker image using provided Dockerfile, or run following command to set additional environment variables: source bashrc_alignsar

Tutorial and sample data

A set of sample data covering the city of Groningen, the Netherlands, with STAC can be downloaded via link-eotdl. More description is in the Tutorial.

Citations

[1] Ling Chang, Anurag Kulshrestha, Bin Zhang and Xu Zhang (2023). Extraction and analysis of radar scatterer attributes for PAZ SAR by combining time series InSAR, PolSAR and land use measurements. Remote Sensing 15(6), 1571 [DOI].

[2] Anurag Kulshrestha, Ling Chang and Alfred Stein (2024). Radarcoding reference data for SAR training data creation in Radar coordinates. IEEE Geoscience and Remote Sensing Letters.

[3] Ling Chang, Jose Manuel Delgado Blasco, Andrea Cavallini, Andy Hooper, Anurag Kulshrestha, Milan Lazecky, Wojciech Witkowski, Xu Zhang, Serkan Girgin and all other AlignSAR members (2023). AlignSAR: developing an open SAR library for machine learning applications. ESA Fringe 2023, 11-15 Sept 2023, Leeds, UK.

[4] Ling Chang, Xu Zhang, Anurag Kulshrestha, Serkan Girgin, Alfred Stein, Jose Manuel Delgado Blasco, Angie Catalina Carrillo Chappe, Andrea Cavallini, Marco Uccelli, Milan Lazecky, Andy Hooper, Wojciech Witkowski, Magdalena Lucka, Artur Guzy (2024). AlignSAR: An open-source package of SAR benchmark dataset creation for machine learning applications. 2024 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2024, accepted.

License

CC BY-NC-SA: [Creative Commons Attribution-NonCommercial-ShareAlike][https://creativecommons.org/licenses/by-nc-sa/4.0/]

This license lets others remix, adapt, and build upon your work non-commercially, as long as they credit you and license their new creations under the identical terms.

Questions and suggestions

We welcome your questions and suggestions and contributions. Free free to contact us by sending email to alignsar.project@gmail.com

Acknowledgment

We acknowledge ESA.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • C++ 41.1%
  • C 25.8%
  • Jupyter Notebook 16.8%
  • Python 10.9%
  • Shell 4.7%
  • Roff 0.4%
  • Other 0.3%