Objective comparison of relief visualization techniques with deep CNN for archaeology

被引:29
作者
Guyot, Alexandre [1 ,2 ]
Lennon, Marc [2 ]
Hubert-Moy, Laurence [1 ]
机构
[1] Univ Rennes 2, Lab LETG UMR 6554, Pl Recteur Henri Le Moal, F-35043 Rennes, France
[2] Hytech Imaging, 115 Rue Claude Chappe, F-29280 Plouzane, France
关键词
Archaeological prospection; Remote-sensing; Computer vision; Deep convolutional network; LiDAR; Airborne laser system; SKY-VIEW FACTOR; MODELS;
D O I
10.1016/j.jasrep.2021.103027
中图分类号
K85 [文物考古];
学科分类号
0601 ;
摘要
Archaeology has been profoundly transformed by the advent of airborne laser scanning (ALS) technology (a.k.a airborne LiDAR). High-resolution and high-precision synoptic views of earth's topography are now available, even in densely forested environments, to identify and characterize landform patterns resulting from past human occupation. ALS-based archaeological prospection relies on digital terrain model (DTM) visualization techniques (VTs) that highlight subtle topographical changes perceived and interpreted by archaeologists. An increasing number of VTs have been developed, and they have been evaluated to date mainly based on subjective human perception. This study developed a new approach based on state-of-the-art computer-vision algorithms to benchmark VTs using objective metrics. Thirteen VTs were applied to a ALS-derived DTM, and a deep convolution neural network (deep CNN) was implemented and trained to automatically detect and segment archaeological structures from these images. Visual interpretation of the images showed that the most informative VT was e2MSTP, which combined a multiscale topographic analysis (MSTP) with a morphologically explicit image and a slope-invariant relief detrending technique. The deep CNN approach confirmed these results and provided objective performance metrics. This study indicates that the computer vision approach opens new perspectives in the objective selection of the most suitable VT for archaeological prospection.
引用
收藏
页数:8
相关论文
共 47 条
[1]  
Abdulla Waleed, 2017, MATTERPORTMASK RCNN
[2]  
[Anonymous], 2019, REMOTE SENS-BASEL, DOI DOI 10.3390/rs11171994
[3]  
[Anonymous], 2016, OCEANS DATA
[4]   A Comparison of Visualization Techniques for Models Created from Airborne Laser Scanned Data [J].
Bennett, Rebecca ;
Welham, Kate ;
Hill, Ross A. ;
Ford, Andrew .
ARCHAEOLOGICAL PROSPECTION, 2012, 19 (01) :41-48
[5]   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 [J].
Bonhage, Alexander ;
Eltaher, Mahmoud ;
Raab, Thomas ;
Breuss, Michael ;
Raab, Alexandra ;
Schneider, Anna .
ARCHAEOLOGICAL PROSPECTION, 2021, 28 (02) :177-186
[6]   Using CNN Features to Better Understand What Makes Visual Artworks Special [J].
Brachmann, Anselm ;
Barth, Erhardt ;
Redies, Christoph .
FRONTIERS IN PSYCHOLOGY, 2017, 8
[7]  
Chiba T., 2008, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci, V37, P1071
[8]   Visualisation of LiDAR terrain models for archaeological feature detection [J].
Devereux, B. J. ;
Amable, G. S. ;
Crow, P. .
ANTIQUITY, 2008, 82 (316) :470-479
[9]   Sensitivity Analysis of k-Fold Cross Validation in Prediction Error Estimation [J].
Diego Rodriguez, Juan ;
Perez, Aritz ;
Antonio Lozano, Jose .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2010, 32 (03) :569-575
[10]  
Doneus M., 2006, BAR INT SERIES, V1568, P99