Tools

  1. geodl: An R package for geospatial deep learning using torch, luz, and terra.
  2. WV LiDAR and Elevation Download Tool: Web app that allows users to download LiDAR data and derived products collected within the state of West Virginia.

Datasets

  1. mineBenchDL: A geomorphology deep learning dataset of historic surface coal mine benches in West Virginia, USA.
  2. mineBenchDL (source): Source digital terrain model and vector mine bench features for mineBenchDL dataset. New land surface parameters can be derived from the digital elevation data.
  3. topoDL: A deep learning semantic segmentation dataset for the extraction of surface mine extents from historic USGS topographic maps.
  4. wvSlpFailureML: A dataset for slope failure occurrence predictive modeling using machine learning and LiDAR -derived topographic variables for the entirety of the state of West Virginia, USA.
  5. surficialDL: A geomorphology deep learning dataset of alluvium and thick glacial till derived from 1:24,000 scale surficial geology data for the western portion of Massachusetts, USA.
  6. terraceDL: A geomorphology deep learning dataset of agricultural terraces in Iowa, USA.
  7. vfillDL: A geomorphology deep learning dataset of valley fill faces resulting from mountaintop removal coal mining (southern West Virginia, eastern Kentucky, and southwestern Virginia, USA)
  8. wvLC2016: 2016 land cover dataset for West Virginia derived from National Agriculture Imagery Program (NAIP) orthophotography using geographic objected-based image analysis and machine learning. This product is associated with the following publication: "Large-Area, High Spatial Resolution Land Cover Mapping using Random Forests, GEOBIA, and NAIP Orthophotography: Findings and Recommendations".
  9. ctWhgtAcc: R-based assessment tool, instructions, and example data associated with "Thematic Classification Accuracy Assessment with Inherently Uncertain Boundaries: An Argument for Center-Weighted Accuracy Assessment Metrics".
  10. wvlcDL_5m: Image chips derived from manually digitized training data and NAIP orthophotography for use in deep learning-based semantic segmentation experiments. This dataset represents an incomplete training set where each cell is not mapped to a land cover class. These data are used in our Geospatial Deep Learning course.
  11. GEE Extract Example: Code example demonstrating how to use the geemap python library to extract pixel values at point locations for a timeseries of images using Google Earth Engine. A video that walks through the process is available here.
  12. slidesEBM: Data and code associated with "Explainable Boosting Machines for Slope Failure Spatial Predictive Modeling".