Virtual Diving Analysis of the archaeological value in image-based digital reconstructions of underwater shipwreck sites in Western Australia
Madeline McAllister
Image: Batavia, WA Museum
PhD Candidate The University of Western Australia
[email protected]
Overview • PhD research (aims, background) • Image-based digital recording and archaeological value • Accuracy • Repeatability • Wide-angle distortion • Movement – fish, seaweed etc
Image: Rockingham Dive Trail, Cessna M. McAllister
Shipwrecks of the Roaring Forties A maritime archaeological reassessment of some of Australia’s earliest shipwrecks (LP130100137, Lead CI: Dr Alistair Paterson)
Problem Statement ¥ Archaeology = destructive and irreversible ¥ Non-intrusive surveys are more prevalent but, there is always change and degradation through natural and human forces
Image: R40s, N. Bigourdan
¥ So our recording must be as thorough and accurate as possible
Background Underwater Photogrammetry First experiments began in the 1960s and have continually evolved and been applied, more recently with particular focus on creating 3D digital site reconstructions Image: Submerged ruins, Pavlopetri, Henderson et Al. 2012
Advances in Technology
• Advance in algorithms for computer science and machine vision • Increasing performance and automation • Affordability and accessibility Images: Agisoft Photoscan logo, Moulan & Bezzi nd:12
Benefits of 3D Recording • • • • • • • • • • •
Cross-comparisons Enhanced archaeological interpretation Digital preservation Monitoring shape and colours Simulation of ageing and deterioration Heritage management Virtual reality/computer graphics applications 3D repositories and catalogues Web-based visualisation Computer aided restoration Multimedia museum exhibitions
(McCArthy 2014, Pierrot-Deseilligny et al 2011) Image: Zeewijk, WA Museum
Theoretical Positioning? Subjective and bias influence archaeological recording and interpretation at every level. How do digital 3D reconstructions affect this?
Henderson et Al 2013:250
“The process of archaeological visualization [sic] refers to the agency of the products …and how they perform as key interpretive and explanatory tools in our discipline. While archaeological visualizations [sic] serve as a critical “gateway” through which disciplinary knowledge can be accessed by non-professionals, they also enable our own professional exploration of ideas and concepts that are difficult to identify and articulate through textual discourse.” (Moser, S. 2012:296)
PhD Research ¥ To develop and apply a methodology for photographically recording an underwater shipwreck site and processing that data (both legacy and new) into a geometrically accurate 3D reconstruction, which allows for archaeological interpretation. ¥ To understand the limits and inaccuracies of underwater photogrammetry ¥ To discuss the archaeological value of image-based, digital recording and where it fits in the overall archaeological process and theoretical debates.
Photographic data – legacy and new
WA Museum: Batavia, P. Baker, WA Museum
Accuracy of what? • Geometrical • Spatial • Colour
“Fast is fine, but accuracy is everything.”
• Texture • Repeatability
- Wyatt Earp
en.wikipedia.org
Reliable Reproducibility Case study: Pool test frame – controlled underwater environment Accuracy in terms of multiple reconstructions and repeatability of a method on a site overtime.
Outcome: able to compare differences in multiple 3D models of the same object Reconstruction software: Agisoft Photoscan (standard) Comparison software: CloudCompare (opensource).
Image: M. McAllister
¥ All images taken with a SONY RX100II in an Ikelite housing ¥ Set of 161 images used to create the Base Comparison model below ¥ Images were taken in groups at slightly different angles
Base Model
Comparison Model 1
Comparison Model 3
Comparison Model 2
Comparison Model 4
CloudCompare
Image: M. McAllister
Reliable and repeatable? The comparison of dense point clouds shows that repeatedly creating models from images taken at similar view points are reliable Models made from the smaller image sets capture essential information (size, form etc). Models from larger image sets (i.e. Base model) give more detail and resolution to the point clouds.
To wide-angle, or, not to wide-angle?
Image: M. McAllister
SonyRX100 2, One metre away
SonyRX100 2 + wide angle lens, 1 metre away
Image: M. McAllister
Larger sites: Batavia (1629)
Image: M. McAllister
Batavia (1629) Legacy data
Images: WA Museum
Camera: SONYRX100II in Ikelite Housing Image set: 142 Site length: Approximately 35m (20m recorded)
Image: M. McAllister
Images: M. McAllister
Batavia (1629) Reconstruction: Feb 2015
Conclusions a) Photoscan software is reliable for repeating 3D digital reconstructions of underwater objects in a controlled environment b) Next stage: testing this on a shipwreck site with more realistic conditions c) Already the Batavia site shows that lots of fish do not pose such a problem in terms of identifying the shipwreck features. d) Stress the importance of recording all relevant information to do with how the data was captured and why.
References
Baker, P.E. and J.N. Green 1976 Recording techniques used during the excavation of the Batavia. International Journal of Nautical Archaeology 5(2):143-158. Bateman, J. 2000 Immediate realities: an anthropology of computer visualisation in archaeology. Internet Archaeology 8. Bateman, J. 2005 Wearing Juninho's shirt: record and negotiation in excavation photographs. Envisioning the past: archaeology and the image:192-203. De Reu, J., G. Plets, G. Verhoeven, P. De Smedt, M. Bats, B. Cherretté, W. De Maeyer, J. Deconynck, D. Herremans, P. Laloo, M. Van Meirvenne and W. De Clercq 2013 Towards a three-dimensional cost-effective registration of the archaeological heritage. Journal of Archaeological Science 40(2):1108-1121. Dechert, P. 1975 Communication, Photography, and the Archaeologist. In E. Harp (ed.), Photogrpahy in Archaeological Research, pp. 345-364. Alburquerque: University of New Mexico Press. Drap, P., D. Merad, J. Seinturier, A. Mahiddine, D. Peloso, J.-M. Boi, B. Chemisky, L. Long and J. Garrabou 2013 Underwater programmetry for archaeology and marine biology: 40 years of experience in Marseille, France. Digital Heritage International Congress (DigitalHeritage), 2013. Eiteljorg, H. 2000 The compelling computer image–a double-edged sword. Internet Archaeology 8. Evans, T.L. and P.T. Daly 2006 Digital archaeology: bridging method and theory: Psychology Press. Frankland, T. and G. Earl 2011 Authority and authenticity in future archaeological visualisation. ADS-VIS, Huddersfield: University of Huddersfield. Frischer, B. 2008 From digital illustration to digital heuristics. Beyond Illustration: 2D and 3D Digital Technologies as Tools for Discovery in Archaeology 1805. Gillings, M. 2005 The real, the virtually real, and the hyperreal: The role of VR in archaeology. Envisioning the Past: Archaeology and the Image:223-239. Gregory, D., P. Jensen and K. Strætkvern 2012 Conservation and in situ preservation of wooden shipwrecks from marine environments. Journal of Cultural Heritage 13(3, Supplement):S139-S148. Harp, E. 1975 Photography in archaeological research: University of New Mexico Press. Hodder, I. 1999 The archaeological process: An introduction: Blackwell Oxford. Huggett, J. 2012 What lies beneath: lifting the lid on archaeological computing. Miller, P. and J. Richards 1995 The good, the bad, and the downright misleading: archaeological adoption of computer visualisation. BAR INTERNATIONAL SERIES 600:19-19. Moser, S. 2012 Early Artifact Illustration and the Birth of the Archaeological Image. Archaeological Theory Today:292. Olson, B.R., R.A. Placchetti, J. Quartermaine and A.E. Killebrew 2013 The Tel Akko Total Archaeology Project (Akko, Israel): Assessing the suitability of multi-scale 3D field recording in archaeology. Journal of Field Archaeology 38(3):244-262. Ware, C. 2004 Information Visualization: Perception for Design. San Francisco: Elsevier Science & Technology Books. Zubrow, E.B. 2006 Digital Archaeology. Digital Archaeology. Bridging method and theory, London:10-31.
