Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA

The logistics, costs, and capacity needed to complete extensive archaeological pedestrian surveys to inventory cultural resources present challenges to public land managers. To address these issues, we developed a workflow combining lidar-derived imagery and deep learning (DL) models tailored for cu...

Full description

Saved in:
Bibliographic Details
Main Authors: Claudine Gravel-Miguel, Grant Snitker, Jayde N. Hirniak, Katherine Peck, Alex Fetterhoff
Format: Article
Language:English
Published: Cambridge University Press
Series:Advances in Archaeological Practice
Subjects:
Online Access:https://www.cambridge.org/core/product/identifier/S2326376825000014/type/journal_article
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849233077702754304
author Claudine Gravel-Miguel
Grant Snitker
Jayde N. Hirniak
Katherine Peck
Alex Fetterhoff
author_facet Claudine Gravel-Miguel
Grant Snitker
Jayde N. Hirniak
Katherine Peck
Alex Fetterhoff
author_sort Claudine Gravel-Miguel
collection DOAJ
description The logistics, costs, and capacity needed to complete extensive archaeological pedestrian surveys to inventory cultural resources present challenges to public land managers. To address these issues, we developed a workflow combining lidar-derived imagery and deep learning (DL) models tailored for cultural resource management (CRM) programs on public lands. It combines Python scripts that fine-tune models to recognize archaeological features in lidar-derived imagery with denoising QGIS steps that improve the predictions’ performance and applicability. We present this workflow through an applied case study focused on detecting historic agricultural terraces in the Piedmont National Wildlife Refuge, Georgia, USA. For this project, we fine-tuned pretrained U-Net models to teach them to recognize agricultural terraces in imagery, identified the parameter settings that led to the highest recall for detecting terraces, and used those settings to train models on incremental dataset sizes, which allowed us to identify the minimum training size necessary to obtain satisfying models. Results present effective models that can detect most terraces even when trained on small datasets. This study provides a robust methodology that requires basic proficiencies in Python coding but expands DL applications in federal CRM by advancing methods in lidar and machine learning for archaeological inventorying, monitoring, and preservation.
format Article
id doaj-art-e09907d46dc44848ba4908a6667976c2
institution Kabale University
issn 2326-3768
language English
publisher Cambridge University Press
record_format Article
series Advances in Archaeological Practice
spelling doaj-art-e09907d46dc44848ba4908a6667976c22025-08-20T13:02:15ZengCambridge University PressAdvances in Archaeological Practice2326-376812410.1017/aap.2025.1Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USAClaudine Gravel-Miguel0https://orcid.org/0000-0003-1324-7937Grant Snitker1https://orcid.org/0000-0003-3548-2241Jayde N. Hirniak2https://orcid.org/0000-0001-6491-8097Katherine Peck3https://orcid.org/0009-0008-4068-2591Alex Fetterhoff4Cultural Resource Sciences Program, New Mexico Consortium, Los Alamos, NM, USA Department of Anthropology, University of New Mexico, Albuquerque, NM, USACultural Resource Sciences Program, New Mexico Consortium, Los Alamos, NM, USA Department of Anthropology, University of New Mexico, Albuquerque, NM, USAInstitute of Human Origins, School of Human Evolution and Social Change, Arizona State University, Tempe, AZ, USACultural Resource Sciences Program, New Mexico Consortium, Los Alamos, NM, USA Department of Anthropology, University of New Mexico, Albuquerque, NM, USACultural Resource Sciences Program, New Mexico Consortium, Los Alamos, NM, USA Department of Anthropology, University of New Mexico, Albuquerque, NM, USAThe logistics, costs, and capacity needed to complete extensive archaeological pedestrian surveys to inventory cultural resources present challenges to public land managers. To address these issues, we developed a workflow combining lidar-derived imagery and deep learning (DL) models tailored for cultural resource management (CRM) programs on public lands. It combines Python scripts that fine-tune models to recognize archaeological features in lidar-derived imagery with denoising QGIS steps that improve the predictions’ performance and applicability. We present this workflow through an applied case study focused on detecting historic agricultural terraces in the Piedmont National Wildlife Refuge, Georgia, USA. For this project, we fine-tuned pretrained U-Net models to teach them to recognize agricultural terraces in imagery, identified the parameter settings that led to the highest recall for detecting terraces, and used those settings to train models on incremental dataset sizes, which allowed us to identify the minimum training size necessary to obtain satisfying models. Results present effective models that can detect most terraces even when trained on small datasets. This study provides a robust methodology that requires basic proficiencies in Python coding but expands DL applications in federal CRM by advancing methods in lidar and machine learning for archaeological inventorying, monitoring, and preservation.https://www.cambridge.org/core/product/identifier/S2326376825000014/type/journal_articleconvolutional neural network (CNN)cultural resource management (CRM)deep learninglidarterracesU-Netred neuronal convolucional (RNC)programas de gestión de recursos culturales (PGRC)aprendizaje profundolidarterrazasU-Net
spellingShingle Claudine Gravel-Miguel
Grant Snitker
Jayde N. Hirniak
Katherine Peck
Alex Fetterhoff
Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA
Advances in Archaeological Practice
convolutional neural network (CNN)
cultural resource management (CRM)
deep learning
lidar
terraces
U-Net
red neuronal convolucional (RNC)
programas de gestión de recursos culturales (PGRC)
aprendizaje profundo
lidar
terrazas
U-Net
title Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA
title_full Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA
title_fullStr Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA
title_full_unstemmed Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA
title_short Semantic Segmentation of Archaeological Features on Public Lands: Case Study of Historical Cotton Terraces within the Piedmont National Wildlife Refuge, Georgia, USA
title_sort semantic segmentation of archaeological features on public lands case study of historical cotton terraces within the piedmont national wildlife refuge georgia usa
topic convolutional neural network (CNN)
cultural resource management (CRM)
deep learning
lidar
terraces
U-Net
red neuronal convolucional (RNC)
programas de gestión de recursos culturales (PGRC)
aprendizaje profundo
lidar
terrazas
U-Net
url https://www.cambridge.org/core/product/identifier/S2326376825000014/type/journal_article
work_keys_str_mv AT claudinegravelmiguel semanticsegmentationofarchaeologicalfeaturesonpubliclandscasestudyofhistoricalcottonterraceswithinthepiedmontnationalwildliferefugegeorgiausa
AT grantsnitker semanticsegmentationofarchaeologicalfeaturesonpubliclandscasestudyofhistoricalcottonterraceswithinthepiedmontnationalwildliferefugegeorgiausa
AT jaydenhirniak semanticsegmentationofarchaeologicalfeaturesonpubliclandscasestudyofhistoricalcottonterraceswithinthepiedmontnationalwildliferefugegeorgiausa
AT katherinepeck semanticsegmentationofarchaeologicalfeaturesonpubliclandscasestudyofhistoricalcottonterraceswithinthepiedmontnationalwildliferefugegeorgiausa
AT alexfetterhoff semanticsegmentationofarchaeologicalfeaturesonpubliclandscasestudyofhistoricalcottonterraceswithinthepiedmontnationalwildliferefugegeorgiausa