This repository contains the custom Python script used for automated Synthetic Aperture Radar (SAR) data processing and Radio Frequency Interference (RFI) hotspot extraction via the Google Earth Engine (GEE) Python API.
Traditional SAR data processing pipelines often treat active electromagnetic interference as "noise" or "artifacts" to be removed. This script repositions C-band Radio Frequency Interference (RFI) as a highly valuable, passive signature of electromagnetic anomalies (such as electronic warfare operations).
The script extract_s1_rfi_global.py executes a dual-threshold statistical detection algorithm directly on GEE servers. It processes massive global archives of Sentinel-1A Ground Range Detected (GRD) imagery, extracts features at a 500m scale leveraging image pyramids, and outputs the geographic coordinates and temporal frequencies of RFI events as lightweight CSV files.
- Triple Pre-masking: Excludes water bodies (via ESA WorldCover), low signal-to-noise ratio areas, and extremely low VH background regions to suppress false alarms.
- Dual-Threshold Feature Extraction: * Intensity Indicator: 99th percentile of VH polarization backscatter (
VH_p99 > -13.0 dB).- Polarization Anomaly: 95th percentile of the Radio Frequency Interference Index (
RFII_p95 > -0.5).
- Polarization Anomaly: 95th percentile of the Radio Frequency Interference Index (
- Spatial Constraint: Requires a minimum contiguous valid footprint (~50 km²) to rule out isolated salt-and-pepper noise.
- A registered and approved Google Earth Engine (GEE) account.
- A local Python environment (Python 3.7+ recommended).
- The Earth Engine Python API installed.
pip install ee