A novel Structure from Motion-based approach to underwater pile field documentation. (October 2021)

• Record Type: Journal Article
• Authors: Reich, Johannes; Steiner, Philipp; Ballmer, Ariane; Emmenegger, Lea; Hostettler, Marco; Stäheli, Corinne; Naumov, Goce; Taneski, Bojan; Todoroska, Valentina; Schindler, Konrad; Hafner, Albert
• Is Part Of: Journal of archaeological science. Volume 39(2021)
• Publication Date: 2021-10
• Subjects: Underwater archaeology -- Prehistoric lakeside settlements -- Pile fields -- 3D-documentation -- Photogrammetry -- Structure from Motion (SfM) -- Deep Convolutional Neural Network; Archaeology -- Periodicals; Archaeology -- Research -- Periodicals; 930.1
• DOI: 10.1016/j.jasrep.2021.103120
• Languages: English
• ISSNs: 2352-409X

Abstract

Highlights: A SfM-based pile field documentation workflow is a effective alternative to conventional methods. SfM yields spatial information in highest resolution at low expenses. A machine learning approach enables the detection and masking of fishes on the photos. Abstract: This article presents a novel methodology to the underwater documentation of pile fields in archaeological lakeside settlement sites using Structure from Motion (SfM). Mapping the piles of such sites is an indispensable basis to the exploitation of the high resolution absolute chronological data gained through dendrochronology. In a case study at the underwater site of Ploča, Mičov Grad at Lake Ohrid, North Macedonia, nine consecutive 10 m 2 strips and a 6 m 2 excavation section were uncovered, the situation documented, and the wood piles sampled. The gained data was vectorized in a geographic information system. During two field campaigns, a total of 794 wooden elements on a surface of 96 m 2 could be documented three-dimensionally with a residual error of less than 2 cm. The exceptionally high number of fishes in the 5 m deep water resulted in a significant covering of potentially important information on the relevant photos. We present a machine learning approach, especially developed and successfully applied to the automatic detection and masking of these fishes in order to eliminate them from the images. The discussed documentation workflow enables an efficient, cost-effective, accurate and reproducible mapping of pile fields. So far, no other method applied to the recording of pile fields has allowed for a comparably high resolution of spatial information.