ARHO land use planning database compatible with national Ryhti APIs

This project is part of the MIT 2023 WORLDING workshop, a research and development initiative exploring climate futures at the intersection of documentary, land-use planning, speculative design, and game-engine technologies.

Trained a deep learning-based system using high-resolution Sentinel-2 imagery and CNNs to classify land into water, vegetation, urban and barren areas with 92.4% validation accuracy.