Machine learning for rapid mapping of archaeological structures made of dry stones – Example of burial monuments from the Khirgisuur culture, Mongolia –
Fabrice Monna
(1)
,
Jérôme Magail
(2)
,
Tanguy Rolland
(1)
,
Nicolas Navarro
(3, 4)
,
Josef Wilczek
(1)
,
Jamiyan-Ombo Gantulga
(5)
,
Yury Esin
(6)
,
Ludovic Granjon
(7)
,
Anne-Caroline Allard
(8)
,
Carmela Chateau-Smith
(9, 10)
1
ARTeHiS -
Archéologie, Terre, Histoire, Sociétés [Dijon]
2 Musée d'Anthropologie préhistorique de Monaco
3 BGS - Biogéosciences [UMR 6282]
4 EPHE - École Pratique des Hautes Études
5 MAS - Mongolian Academy of Sciences
6 Institut de Recherche de Khakassie sur les Langues, la Littérature et l’Histoire
7 GEOBFC - Plateforme GEOBFC (Géomatique Bourgogne Franche-Comté)
8 CAC - Centre André Chastel : Laboratoire de recherche en histoire de l'art
9 UFR SVTE - UFR Sciences de la Vie, de la Terre et de l'Environnement (Université de Bourgogne)
10 CPTC - Centre Pluridisciplinaire Textes et Cultures [Dijon]
2 Musée d'Anthropologie préhistorique de Monaco
3 BGS - Biogéosciences [UMR 6282]
4 EPHE - École Pratique des Hautes Études
5 MAS - Mongolian Academy of Sciences
6 Institut de Recherche de Khakassie sur les Langues, la Littérature et l’Histoire
7 GEOBFC - Plateforme GEOBFC (Géomatique Bourgogne Franche-Comté)
8 CAC - Centre André Chastel : Laboratoire de recherche en histoire de l'art
9 UFR SVTE - UFR Sciences de la Vie, de la Terre et de l'Environnement (Université de Bourgogne)
10 CPTC - Centre Pluridisciplinaire Textes et Cultures [Dijon]
Fabrice Monna
- Fonction : Auteur
- PersonId : 9891
- IdHAL : fabrice-monna
- ORCID : 0000-0001-7771-2434
- IdRef : 183448707
Jérôme Magail
- Fonction : Auteur
- PersonId : 179286
- IdHAL : jerome-magail
- ORCID : 0000-0002-5768-0698
- IdRef : 068931409
Nicolas Navarro
- Fonction : Auteur
- PersonId : 759148
- IdHAL : nco-navarro
- ORCID : 0000-0001-5694-4201
- IdRef : 080279155
Josef Wilczek
- Fonction : Auteur
- PersonId : 10789
- IdHAL : josef-wilczek
- ORCID : 0000-0001-8633-5095
Ludovic Granjon
- Fonction : Auteur
- PersonId : 5368
- IdHAL : ludovic-granjon
- ORCID : 0000-0002-0576-5497
- IdRef : 253134528
Anne-Caroline Allard
- Fonction : Auteur
- PersonId : 1214951
- IdHAL : anne-caroline-allard
- ORCID : 0000-0003-3578-3829
Carmela Chateau-Smith
- Fonction : Auteur
- PersonId : 759727
- ORCID : 0000-0001-6995-1670
- IdRef : 168393514
Résumé
The present study proposes a workflow to extract from orthomosaics the enormous amount of dry stones used by past societies to construct funeral complexes in the Mongolian steppes. Several different machine learning algorithms for binary pixel classification (i.e. stone vs non-stone) were evaluated. Input features were extracted from high-resolution orthomosaics and digital elevation models (both derived from aerial imaging). Comparative analysis used two colour spaces (RGB and HSV), texture features (contrast, homogeneity and entropy raster maps), and the topographic position index, combined with nine supervised learning algorithms (nearest centroid, naive Bayes, k-nearest neighbours, logistic regression, linear and quadratic discriminant analyses, support vector machine, random forest, and artificial neural network). When features are processed together, excellent output maps, very close to or outperforming current standards in archaeology, are observed for almost all classifiers. The size of the training set can be drastically reduced (to ca. 300 samples) by majority voting, while maintaining performance at the highest level (about 99.5% for all performance scores). Note, however, that if the training set is inadequate or not fully representative, the classification results are poor. That said, the methods applied and tested here are extremely rapid. Extensive mapping, which would have been difficult with traditional, manual, or semi-automatic delineation of stones using a vector graphics editor, now becomes possible. This workflow generally surpasses pedestrian surveys using differential GPS or a total station.
Domaines
Sciences cognitives Sciences de l'environnement Sciences du Vivant [q-bio] Sciences de l'Homme et SociétéFormat du dépôt | Fichier |
---|---|
Type de dépôt | Article dans une revue |
Titre |
en
Machine learning for rapid mapping of archaeological structures made of dry stones – Example of burial monuments from the Khirgisuur culture, Mongolia –
|
Résumé |
en
The present study proposes a workflow to extract from orthomosaics the enormous amount of dry stones used by past societies to construct funeral complexes in the Mongolian steppes. Several different machine learning algorithms for binary pixel classification (i.e. stone vs non-stone) were evaluated. Input features were extracted from high-resolution orthomosaics and digital elevation models (both derived from aerial imaging). Comparative analysis used two colour spaces (RGB and HSV), texture features (contrast, homogeneity and entropy raster maps), and the topographic position index, combined with nine supervised learning algorithms (nearest centroid, naive Bayes, k-nearest neighbours, logistic regression, linear and quadratic discriminant analyses, support vector machine, random forest, and artificial neural network). When features are processed together, excellent output maps, very close to or outperforming current standards in archaeology, are observed for almost all classifiers. The size of the training set can be drastically reduced (to ca. 300 samples) by majority voting, while maintaining performance at the highest level (about 99.5% for all performance scores). Note, however, that if the training set is inadequate or not fully representative, the classification results are poor. That said, the methods applied and tested here are extremely rapid. Extensive mapping, which would have been difficult with traditional, manual, or semi-automatic delineation of stones using a vector graphics editor, now becomes possible. This workflow generally surpasses pedestrian surveys using differential GPS or a total station.
|
Auteur(s) |
Fabrice Monna
1
, Jérôme Magail
2
, Tanguy Rolland
1
, Nicolas Navarro
3, 4
, Josef Wilczek
1
, Jamiyan-Ombo Gantulga
5
, Yury Esin
6
, Ludovic Granjon
7
, Anne-Caroline Allard
8
, Carmela Chateau-Smith
9, 10
1
ARTeHiS -
Archéologie, Terre, Histoire, Sociétés [Dijon]
( 184759 )
- Université de Bourgogne - 6, Bd. Gabriel - 21000 Dijon
- France
2
Musée d'Anthropologie préhistorique de Monaco
( 159375 )
- 56bis bd du Jardin Exotique
- Monaco
3
BGS -
Biogéosciences [UMR 6282]
( 182913 )
- Université de Bourgogne - 6 boulevard Gabriel - 21000 Dijon
- France
4
EPHE -
École Pratique des Hautes Études
( 110691 )
- 4-14 Rue Ferrus, 75014 Paris
- France
5
MAS -
Mongolian Academy of Sciences
( 337004 )
- Amariin gudamj 1, Ulaanbaatar 210620, Mongolia
- Mongolie
6
Institut de Recherche de Khakassie sur les Langues, la Littérature et l’Histoire
( 506250 )
- 23 Shchetinkin Str. Abakan, 655017 République de
Khakassie, Russie
- Russie
7
GEOBFC -
Plateforme GEOBFC (Géomatique Bourgogne Franche-Comté)
( 253432 )
- MSHD - 6 Esplanade Erasme BP 26611 21066 Dijon cedex
- France
8
CAC -
Centre André Chastel : Laboratoire de recherche en histoire de l'art
( 542230 )
- Galerie Colbert - INHA 2e étage 2 rue Vivienne 75002 PARIS
- France
9
UFR SVTE -
UFR Sciences de la Vie, de la Terre et de l'Environnement (Université de Bourgogne)
( 210093 )
- 6 boulevard Gabriel - 21000 Dijon
- France
10
CPTC -
Centre Pluridisciplinaire Textes et Cultures [Dijon]
( 496953 )
- Université de Bourgogne - Faculté des Lettres et philosophie - 2 bd Gabriel - 21000 DIJON
- France
|
Licence |
Paternité - Pas d'utilisation commerciale
|
Langue du document |
Anglais
|
Nom de la revue |
|
Vulgarisation |
Non
|
Comité de lecture |
Oui
|
Audience |
Internationale
|
Date de publication |
2020-01
|
Volume |
43
|
Page/Identifiant |
118-128
|
URL éditeur |
https://www-sciencedirect-com/science/article/pii/S1296207419305333?via%3Dihub
|
Commentaire |
11 pages
|
Domaine(s) |
|
Référence interne |
|
Collaboration/Projet |
|
Financement |
|
Mots-clés |
en
Pixel classification, Grey level co-occurrence matrix, RGB colour space, Texture, Topographic position index, Photogrammetry, Burial complex planigraphy, Mongolia, Bronze age, Iron age
|
DOI | 10.1016/j.culher.2020.01.002 |
PII | S1296-2074(19)30533-3 |
Origine :
Fichiers produits par l'(les) auteur(s)
Loading...