TY - JOUR
T1 - Machine learning for automatic detection of historic stone walls using LiDAR data
AU - Trotter, Ezra Francis Leslie
AU - Fernandes, Ana Cristina Mosebo
AU - Fibæk, Casper Samsø
AU - Keßler, Carsten
N1 - Funding Information:
The present manuscript is based on the dissertation entitled ?Machine Learning for Automatic Detection of Historic Stone Walls Using LiDAR Data?, produced by the authors E. F. L. T. and A. C. M. F. (Fernandes and Trotter2021).
Publisher Copyright:
© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
PY - 2022
Y1 - 2022
N2 - Stone walls in the landscape of Denmark are protected not only for their cultural and historical significance but also for their vital role in supporting local biodiversity. Many stone wall structures have either disappeared, suffered substantial damage, or had segments removed. Additionally, as it stands today, the registry of these structures, managed by each municipality, is outdated and incomplete. Leveraging recent developments in Machine Learning and Convolutional Neural Networks (CNNs), we analyze the publicly available terrain data (40 cm resolution) derived from the Danish LiDAR data, using a U-Net-like CNN model to assess the stone walls dataset and provide for an update of the registry. While the Digital Terrain Model (DTM) alone provided good results, better results were obtained when adding Height Above Terrain (HAT) and an additional DTM layer with a Sobel filter applied. Using a pixel-wise evaluation, there was an overall agreement of 93% between ground truth and prediction of stone walls in a validation area and 88% overall agreement for the whole predicted area. Good generalizability was found when externally validating the model on new data, showing positive results for both the existing stone walls and predicting new potential ones upon visualisation. The method performed best in open areas, however positive results were also seen in forested areas, although denser areas and urban areas presented as challenging. Given the lack of a reference dataset or other studies on this specific matter, the evaluation of our study was heavily based on the stone walls registry itself complemented by visual inspection of the predictions and on the ground in the Danish municipality of Ærø. Automating the process of identifying and updating the stone walls registry in Denmark is of great relevance to the local governments. We suggest the development of a Decision Support System to allow municipalities access to the results of this method.
AB - Stone walls in the landscape of Denmark are protected not only for their cultural and historical significance but also for their vital role in supporting local biodiversity. Many stone wall structures have either disappeared, suffered substantial damage, or had segments removed. Additionally, as it stands today, the registry of these structures, managed by each municipality, is outdated and incomplete. Leveraging recent developments in Machine Learning and Convolutional Neural Networks (CNNs), we analyze the publicly available terrain data (40 cm resolution) derived from the Danish LiDAR data, using a U-Net-like CNN model to assess the stone walls dataset and provide for an update of the registry. While the Digital Terrain Model (DTM) alone provided good results, better results were obtained when adding Height Above Terrain (HAT) and an additional DTM layer with a Sobel filter applied. Using a pixel-wise evaluation, there was an overall agreement of 93% between ground truth and prediction of stone walls in a validation area and 88% overall agreement for the whole predicted area. Good generalizability was found when externally validating the model on new data, showing positive results for both the existing stone walls and predicting new potential ones upon visualisation. The method performed best in open areas, however positive results were also seen in forested areas, although denser areas and urban areas presented as challenging. Given the lack of a reference dataset or other studies on this specific matter, the evaluation of our study was heavily based on the stone walls registry itself complemented by visual inspection of the predictions and on the ground in the Danish municipality of Ærø. Automating the process of identifying and updating the stone walls registry in Denmark is of great relevance to the local governments. We suggest the development of a Decision Support System to allow municipalities access to the results of this method.
KW - archaeology
KW - deep learning
KW - DEM
KW - LiDAR
KW - Stone walls
UR - http://www.scopus.com/inward/record.url?scp=85128350797&partnerID=8YFLogxK
U2 - 10.1080/01431161.2022.2057206
DO - 10.1080/01431161.2022.2057206
M3 - Journal article
AN - SCOPUS:85128350797
SN - 0143-1161
VL - 43
SP - 2185
EP - 2211
JO - International Journal of Remote Sensing
JF - International Journal of Remote Sensing
IS - 6
ER -