This repository is an implementation of MineSegSAT, presented at ECRS 2023. MineSegSAT is a deep learning model that identifies environmentally impacted areas of mineral extraction sites using the SegFormer model architecture trained on Sentinel-2 data.
This project uses poetry as a package manager which can be installed by following the instructions found here.
To install the dependencies in this project run the following command while your virtual environment is activated:
poetry installData that accompanies this repository can be downloaded from here. A sample configuration file that could be used for training a model on the provided dataset can be found in the configs directory.