Pre-Processed Dual-Pol Sentinel-1 SAR Dataset for Machine Learning-Based Burned Area Mapping

This dataset consists of pre-processed dual-polarization Sentinel-1 SAR data prepared for machine learning-based burned area mapping in the Palisades, Los Angeles region. The data includes VV and VH backscatter from ascending and descending orbits, as well as derived GLCM texture features computed for both pre- and post-fire periods. Additionally, it provides the difference Normalized Burn Ratio (dNBR) for preparing refernce map and an RGB composite for visualization.

The code for burned area mapping using machine learning with this dataset is available at: https://github.com/rabinatwayana/SAR-Burnt-Area-Mapping-ML.