Data and Code for "Locating Charcoal Production Sites in Sweden Using LiDAR, Hydrological Algorithms, and Deep Learning"

Over the past several centuries, the iron industry played a central role in the economy of Sweden and much of northern Europe. A crucial component of iron manufacturing was the production of charcoal, which was often created in charcoal pits (or piles). These features are visible in LiDAR datasets. These charcoal pits vary in their morphology by region, and training data for some feature types is severely lacking. Here, we investigate the potential for machine automation to aid archaeologists in recording charcoal pit sites with limited training data availability in a forested region of Jönköping County, Sweden. We first use hydrological depression algorithms to conduct a preliminary assessment of the study region and compile suitable training data for charcoal pits. Then, we use these datasets to train a series of RetinaNet deep learning models, which are less computationally expensive than many popular deep learning architectures (e.g., R-CNNs), al-lowing for greater usability. Altogether, our results demonstrate how charcoal production sites can be automatically extracted from lidar datasets, which has great implications for improving our understanding of long-term environmental impact of the iron industry across Northern Europe. Furthermore, our workflow for developing and implementing deep learning models for archaeological research can expand the use of such methods to regions that lack suitable training data.

Citation

Davis, Dylan; Lundin, Julius (2021). Data and Code for "Locating Charcoal Production Sites in Sweden Using LiDAR, Hydrological Algorithms, and Deep Learning" [Data set]. Scholarsphere. https://doi.org/10.26207/zfc5-rd31

Files

Metadata

Work Title Data and Code for "Locating Charcoal Production Sites in Sweden Using LiDAR, Hydrological Algorithms, and Deep Learning"
Access
Open Access
Creators
  1. Dylan Davis
  2. Julius Lundin
Keyword
  1. Charcoal piles
  2. Feep learning
  3. RetinaNet
  4. Hydrological depression analysis
  5. LiDAR
  6. Sweden
License CC BY 4.0 (Attribution)
Work Type Dataset
Publication Date 2021
DOI doi:10.26207/zfc5-rd31
Related URLs
Deposited August 18, 2021

Versions

Analytics

Collections

This resource is currently not in any collection.

Work History

Version 1
published

  • Created
  • Added Creator Dylan Davis
  • Added Creator Julius Lundin
  • Added ResNet151_Model.zip
  • Updated License Show Changes
    License
    • https://creativecommons.org/licenses/by/4.0/
  • Published
  • Updated
  • Updated

Version 2
published

  • Created
  • Added README.md
  • Published
  • Updated
  • Updated

Version 3
published

  • Created
  • Deleted README.md
  • Added README.txt
  • Published
  • Updated
  • Updated
  • Updated Work Title Show Changes
    Work Title
    • Locating Charcoal Production Sites in Sweden Using LiDAR, Hydrological Algorithms, and Deep Learning
    • Data and Code for "Locating Charcoal Production Sites in Sweden Using LiDAR, Hydrological Algorithms, and Deep Learning"
  • Updated Keyword, Related URLs Show Changes
    Keyword
    • Charcoal piles, Feep learning, RetinaNet, Hydrological depression analysis, LiDAR, Sweden
    Related URLs
    • https://doi.org/10.3390/rs13183680