Plastic-Greenhouse-Mapping-Coupling-Prior-Knowledge-and-Deep-Learning-PGMpkdl

This dataset accompanies the repository PGMpk-dl and the article "A novel approach: Coupling prior knowledge and deep learning methods for large-scale plastic greenhouse extraction using Sentinel-1/2 data". The dataset contains: Sentinel-2 imagery for the study regions. Ground truth labels derived from prior knowledge and manual annotation, provided in GeoTIFF format. Corresponding RGB composites generated from Sentinel-2 bands. Please use the linked GitHub code to complete the full processing and training pipeline. The dataset is organized by region, and designed for research on plasticulture mapping, agricultural monitoring, and remote sensing-based land cover analysis.