Mini journal logo  Home Issue Contents All Issues

What Should We Do With These? Challenges related to (semi-)automatically detected sites and features. A note

Niko Anttiroiko

Cite this as: Anttiroiko, N. 2024 What Should We Do With These? Challenges related to (semi-)automatically detected sites and features. A note, Internet Archaeology 66.


Recent advances in machine learning and computer vision techniques have brought (semi-)automatic feature detection within reach of an increasing number of archaeologists and archaeological institutions, including those in Finland. These techniques improve our ability to detect and gather information on archaeological cultural heritage over vast areas in a highly efficient manner. However, the widespread adoption of such methods can also pose significant challenges for archaeological cultural heritage management, especially in relation to certain types of near-ubiquitous archaeological remains from the 17th-20th centuries.

four aerial views of semi-automatically detected tar kilns and labels
Examples of semi-automatically detected tar kilns and labels used for training the deep learning model. The yellow outline (A) shows a manually created label. Areas highlighted in red (B, C & D) indicate tar kilns predicted by the deep learning model. Airborne laser scanning (ALS) visualisations are based on ALS 5p data from the National Land Survey of Finland 2020

In general, machine learning based methods are especially well suited for detecting features that have relatively uniform characteristics, are present in sufficiently high numbers, and are easily discernible in remote-sensing datasets. In Finland, most archaeological features that meet these criteria are relatively recent features, such as tar and charcoal kilns from the 17th-20th centuries or remains of World War I and II era defensive structures. Although there are some exceptions to this rule, such as prehistoric pitfall trap systems, the archaeological features selected by the use of (semi-)automatic detections is heavily skewed towards only a handful of relatively recent feature types. While this is not necessarily a problem per se, it poses a series of questions for the cultural heritage management sector, such as:

  • How to manage extremely large numbers of relatively standardised sites and features? Should everything be protected? If not, which features should be selected for protection?
  • How to make most efficient use of the data from automatic feature detection?
  • How to verify automatically detected features? Is a GIS-based assessment enough? When is ground truthing required?
  • How to ensure that the (semi-)automatically detected sites and features do not drain resources or divert attention from other kinds of archaeological heritage?

In Finland, these questions have sparked active discussion in response to the results from the LIDARK-project (2021-2022) which focused on automatic detection of archaeological features, especially within the context of ongoing efforts to modernise legislation on the management of archaeological cultural heritage. This article seeks to summarise and reflect on some of these perspectives presented in the Finnish discussion.

Corresponding author: Niko Anttiroiko
Finnish Heritage Agency

Full text

Figure 1: Examples of semi-automatically detected tar kilns and labels used for training the deep learning model

Figure 2: Impact of semi-automated feature detection on the number of known tar kilns in one of the research areas studied in the LIDARK-project

Anttiroiko, N., Groesz, F.J., Ikäheimo, J., Kelloniemi, A., Nurmi, R., Rostad, S. and Seitsonen, O. 2023 'Detecting the archaeological traces of tar production kilns in the northern boreal forests based on airborne laser scanning and deep learning', Remote Sensing 15(7), 1799.

Bonhage, A., Eltaher, M., Raab, T., Breuß, M., Raab, A. and Schneider, A. 2021 'A modified Mask region-based convolutional neural network approach for the automated detection of archaeological sites on high-resolution light detection and ranging-derived digital elevation models in the North German Lowland', Archaeological Prospection 28(2), 177-86.

Davis, D.S. and Lundin, J. 2021 'Locating charcoal production sites in Sweden using LiDAR, hydrological algorithms, and deep learning', Remote Sensing 13(18), 3680.

Ikäheimo, J. 2021 'Tervahautojen ilmalaserkeilausavusteinen työpöytäinventointi Suomussalmella', Muinaistutkija 2021(3), 2-15.

Snitker, G., Moser, J.D., Southerlin, B. and Stewart, C. 2022 'Detecting historic tar kilns and tar production sites using high-resolution, aerial LiDAR-derived digital elevation models: introducing the Tar Kiln Feature Detection workflow (TKFD) using open-access R and FIJI software', Journal of Archaeological Science: Reports 41, 103340.

Suh, J.W., Anderson, E., Ouimet, W., Johnson, K.M. and Witharana, C. 2021 ' Mapping relict charcoal hearths in New England using deep convolutional neural networks and LiDAR data', Remote Sensing 13(22), 4630.

Trier, Ø.D., Reksten, J.H. and Løseth, K. 2021 'Automated mapping of cultural heritage in Norway from airborne lidar data using faster R-CNN', International Journal of Applied Earth Observation and Geoinfromation, 95, 102241.

Verschoof-van der Vaart, W.B. and Lambers, K. 2019 'Learning to look at LiDAR: the use of R-CNN in the automated detection of archaeological objects in LiDAR data from the Netherlands', Journal of Computer Applications in Archaeology 2(1), 31-40.

Internet Archaeology is an open access journal based in the Department of Archaeology, University of York. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

Terms and Conditions | Legal Statements | Privacy Policy | Cookies Policy | Citing Internet Archaeology

Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.