Field systems and later prehistoric land use: New insights into land use detectability and palaeodemography in the Netherlands through LiDAR, automatic detection and traditional field data

Arnoldussen, Stijn; Vaart, Wouter B. Verschoof-van der; Kaptijn, Eva; Bourgeois, Quentin

doi:10.1002/arp.1891

Cited by 3 publications

(1 citation statement)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ultimately, all AFE studies will be incomplete, as it is impossible (and often unnecessary) to identify every single feature in an area. To the contrary, many studies can lead to significant advances in archaeological knowledge even without high levels of accuracy (see Arnoldussen et al, 2022; Verschoof‐van der Vaart & Lambers, 2022).…”

Section: Discussionmentioning

confidence: 99%

EvaluatingMaskR‐CNN models to extract terracing across oceanic high islands: A case study from Sāmoa

Quintus

Davis

Cochrane

2023

Archaeological Prospection

View full text Add to dashboard Cite

Lidar datasets have been crucial for documenting the scale and nature of human ecosystem engineering and land use. Automated analysis methods, which have been rising in popularity and efficiency, allow for systematic evaluations of vast landscapes. Here, we use a Mask R‐CNN deep learning model to evaluate terracing—artificially flattened areas surrounded by steeper slopes—on islands in American Sāmoa. Mask R‐CNN is notable for its ability to simultaneously perform detection and segmentation tasks related to object recognition, thereby providing robust datasets of both geographic locations of terracing features and their spatial morphometry. Using training datasets from across American Sāmoa, we train this model to recognize terracing features and then apply it to the island of Tutuila to undertake an island‐wide survey for terrace locations, distributions and morphologies. We demonstrate that this model is effective (F1 = 0.718), but limitations are also documented that relate to the quality of the lidar data and the size of terracing features. Our data show that the islands of American Sāmoa display shared patterns of terracing, but the nature of these patterns are distinct on Tutuila compared with the Manu'a island group. These patterns speak to the different interior configurations of the islands. This study demonstrates how deep learning provides a better understanding of landscape construction and behavioural patterning on Tutuila and has the capacity to expand our understanding of these processes on other islands beyond our case study.

show abstract

Section: Discussionmentioning

confidence: 99%

EvaluatingMaskR‐CNN models to extract terracing across oceanic high islands: A case study from Sāmoa

Quintus

Davis

Cochrane

2023

Archaeological Prospection

View full text Add to dashboard Cite

show abstract

Assessing the quality of citizen science in archaeological remote sensing: results from the Heritage Quest project in the Netherlands

Bourgeois,

Kaptijn,

Verschoof-van der Vaart

et al. 2024

Antiquity

View full text Add to dashboard Cite

Volunteers are a key part of the archaeological labour force and, with the growth of digital datasets, these citizen scientists represent a vast pool of interpretive potential; yet, concerns remain about the quality and reliability of crowd-sourced data. This article evaluates the classification of prehistoric barrows on lidar images of the central Netherlands by thousands of volunteers on the Heritage Quest project. In analysing inter-user agreement and assessing results against fieldwork at 380 locations, the authors show that the probability of an accurate barrow identification is related to volunteer consensus in image classifications. Even messy data can lead to the discovery of many previously undetected prehistoric burial mounds.

show abstract

The Application of Artificial Intelligence and Machine Learning in Archaeology: Methodological Challenges and a Case Study from Poland

Jakubczak,

Tyszkiewicz,

Leloch

2024

Preprint

View full text Add to dashboard Cite

The use of artificial intelligence (AI) and machine learning (ML) in archaeology has rapidly gained momentum due to its potential to automate and enhance the efficiency of data analysis. This paper examines the application of neural networks for processing Digital Terrain Model (DTM) data to detect archaeological sites in Poland. The study focuses on identifying trenches, mounds, and charcoal kilns through automated image segmentation, utilizing a U-Net convolutional neural network. While the results demonstrate the promise of AI in improving archaeological site detection, various challenges are highlighted, including data quality, feature misclassification, and regional variability in model performance. Additionally, the research underscores the need for interdisciplinary collaboration, as successful implementation requires expertise in archaeology, geospatial analysis, and programming. Despite AI’s potential for time-saving, significant effort is required to ensure accurate annotations and avoid over-reliance on the technology. The case study provides valuable insights into the complexities of applying AI to large-scale archaeological datasets and raises important questions about the limitations and future development of these methods. This paper concludes that while AI offers powerful tools for archaeological research, careful consideration must be given to its methodological challenges. Standardizing procedures and improving data quality will be critical to fully exploiting AI's potential in archaeology.

show abstract

Field systems and later prehistoric land use: New insights into land use detectability and palaeodemography in the Netherlands through LiDAR, automatic detection and traditional field data

Cited by 3 publications

References 54 publications

EvaluatingMaskR‐CNN models to extract terracing across oceanic high islands: A case study from Sāmoa

EvaluatingMaskR‐CNN models to extract terracing across oceanic high islands: A case study from Sāmoa

Assessing the quality of citizen science in archaeological remote sensing: results from the Heritage Quest project in the Netherlands

The Application of Artificial Intelligence and Machine Learning in Archaeology: Methodological Challenges and a Case Study from Poland

Contact Info

Product

Resources

About