Quad-Polarization Oil Spill Detection Dataset (QPOSD)

The Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) dataset focuses on oil spill detection in the Gulf of Mexico, leveraging synthetic aperture radar (SAR) technology. This dataset comprises .mat files containing high-resolution SAR images, paired with corresponding .png label images that provide pixel-level annotations. It categorizes the data into four classes: oil spill, sea surface, lookalikes and land. The dataset is designed to support researchers in developing and validating deep learning models for oil spill detection and classification, enabling more effective environmental monitoring and disaster response. The inclusion of diverse scenarios across the Gulf of Mexico ensures the dataset’s robustness and applicability to real-world challenges.