Machine learning for rapid mapping of archaeological structures made of dry stones – Example of burial monuments from the Khirgisuur culture, Mongolia –
| Autor: | Nicolas Navarro, Jamiyan-Ombo Gantulga, Tanguy Rolland, Josef Wilczek, Carmela Chateau-Smith, Anne-Caroline Allard, Jérôme Magail, Yury Esin, Fabrice Monna, Ludovic Granjon |
|---|---|
| Přispěvatelé: | Archéologie, Terre, Histoire, Sociétés [Dijon] (ARTeHiS), Ministère de la Culture et de la Communication (MCC)-Université de Bourgogne (UB)-Centre National de la Recherche Scientifique (CNRS), Musée d'Anthropologie préhistorique de Monaco, Monaco, Equipe de recherche de Lyon en sciences de l'information et de la communication (ELICO), Sciences Po Lyon - Institut d'études politiques de Lyon (IEP Lyon), Université de Lyon-Université de Lyon-École nationale supérieure des sciences de l'information et des bibliothèques (ENSSIB), Université de Lyon-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université Jean Moulin - Lyon 3 (UJML), Université de Lyon-Université Lumière - Lyon 2 (UL2), Mongolian Academy of Sciences (MAS), Institut de Recherche de Khakassie sur les Langues, la Littérature et l’Histoire, Plateforme GEOBFC (Géomatique Bourgogne Franche-Comté) (GEOBFC), Maison des Sciences de l'Homme de Dijon (MSH Dijon (MSHD)), Université de Bourgogne (UB)-Université Bourgogne Franche-Comté [COMUE] (UBFC)-Centre National de la Recherche Scientifique (CNRS)-Université de Bourgogne (UB)-Université Bourgogne Franche-Comté [COMUE] (UBFC)-Centre National de la Recherche Scientifique (CNRS), UFR Sciences de la Vie, de la Terre et de l'Environnement (Université de Bourgogne) (UFR SVTE), Université de Bourgogne (UB), Research funded by the Join mission Mongolia –Monaco, and the project ROSAS (uB-FC and RNMSH)., ROSAS, Musée d'anthropologie préhistorique de Monaco, École pratique des hautes études (EPHE), Université Paris sciences et lettres (PSL), Biogéosciences [UMR 6282] [Dijon] (BGS), Université de Bourgogne (UB)-AgroSup Dijon - Institut National Supérieur des Sciences Agronomiques, de l'Alimentation et de l'Environnement-Centre National de la Recherche Scientifique (CNRS), Ústav archeologie a muzeologie, Masarykova univerzita, Bronze and Early Iron Age Department, Institut d'Art et d'Archéologie : École doctorale d'Archéologie, Université Panthéon-Sorbonne (UP1), EPHE PSL Research University (EPHE PSL), Université de Bourgogne (UB)-Centre National de la Recherche Scientifique (CNRS) |
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
Archeology
010504 meteorology & atmospheric sciences [SHS.ARCHEO]Humanities and Social Sciences/Archaeology and Prehistory Computer science Materials Science (miscellaneous) Topographic position index [SDV]Life Sciences [q-bio] Conservation Machine learning computer.software_genre 01 natural sciences [SHS]Humanities and Social Sciences Naive Bayes classifier Vector graphics Pixel classification [SCCO]Cognitive science Pixel classification Grey level co-occurrence matrix RGB colour space Texture Topographic position index Photogrammetry Burial complex planigraphy Mongolia Bronze age Iron age 0601 history and archaeology Texture Spectroscopy RGB colour space 0105 earth and related environmental sciences Bronze age 060102 archaeology Artificial neural network business.industry Iron age Centroid Grey level co-occurrence matrix 06 humanities and the arts computer.file_format Mongolia Archaeology Random forest Support vector machine Photogrammetry Chemistry (miscellaneous) [SDE]Environmental Sciences Burial complex planigraphy Artificial intelligence Raster graphics business General Economics Econometrics and Finance computer |
| Zdroj: | Journal of Cultural Heritage Journal of Cultural Heritage, Elsevier, 2020, 43, pp.118-128. ⟨10.1016/j.culher.2020.01.002⟩ |
| ISSN: | 1296-2074 |
| Popis: | 11 pages; International audience; 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect