Programs and Code for Subsurface and MultiSite Geothermal Exploration Artificial Intelligence

This dataset provides Python scripts supporting both subsurface and surface geothermal exploration AI models developed for the project "Detection of Potential Geothermal Exploration Sites from Hyperspectral Images via Deep Learning." It includes two main components: (1) scripts for subsurface geothermal exploration, which generate 3D models from input voxels to predict geothermal potential in subsurface regions, and (2) scripts for multisite surface exploration, utilizing land surface temperature and mineral markers to classify and map geothermal sites on the surface. The dataset covers all stages of processing, from data import and preprocessing through AI model training, testing, and validation, as well as final mapping of geothermal potential areas.

The subsurface exploration scripts generate a 3D geothermal model, while the multisite surface scripts support a 2D classification map from raster input. Requirements include Python 3, TensorFlow 2.4, and a machine with GPU support.