OpenSAR Insight

Welcome to OpenSAR project

OpenSAR is an early-phase project developing ML-ready datasets and algorithms for direct insight generation from raw SAR data. This is a first step toward efficient onboard implementation of end-to-end SAR inference pipelines for low-latency applications.

We compiled novel datasets for three use cases: vessel detection, water body detection, and radio frequency interference (RFI) detection. These datasets include co-registered L0, L1 SLC, and L1 GRD patches with labels partially derived from SARFish/xView3, Kuro Siwo, and Aresys, respectively.

For each use case, we developed two families of ML models: a large, unrestricted model for ground-based use, and a small, optimised model designed for onboard deployment. Datasets and code will be released open source to benefit the research community.

This project has been funded and supported by ESA’s Φ-lab.

Code data

Python and Matlab scripts for database use.

Consortium

Meet the team that developed the project.

Publications

Publications about the project.