A Comprehensive Database of River Obstructions in China

This study combines deep learning and remote sensing imagery to identify various types of artificial river obstructions across different basins in China, including dams, low-head dams, locks, and partial dams. The final dataset includes over 10,000 records of artificial river obstructions, validated through detailed manual visual inspection using Google Maps. Notably, the vector format data currently available on the platform consists of different samples derived from 30% stratified sampling, rather than the entire dataset.