From Deposit to Point Cloud – a Study of Low-Cost Computer Vision Approaches for the Straightforward Documentation of Archaeological Excavations

Abstract

Stratigraphic archaeological excavations demand high-resolution documentation techniques for 3D recording. Today, this is typically accomplished using total stations or terrestrial laser scanners. This paper demonstrates the potential of another technique that is low-cost and easy to execute. It takes advantage of software using Structure from Motion (SfM) algorithms, which are known for their ability to reconstruct camera pose and threedimensional scene geometry (rendered as a sparse point cloud) from a series of overlapping photographs captured by a camera moving around the scene. When complemented by stereo matching algorithms, detailed 3D surface models can be built from such relatively oriented photo collections in a fully automated way. The absolute orientation of the model can be derived by the manual measurement of control points. The approach is extremely flexible and appropriate to deal with a wide variety of imagery, because this computer vision approach can also work with imagery resulting from a randomly moving camera (i.e. uncontrolled conditions) and calibrated optics are not a prerequisite. For a few years, these algorithms are embedded in several free and low-cost software packages. This paper will outline how such a program can be applied to map archaeological excavations in a very fast and uncomplicated way, using imagery shot with a standard compact digital camera (even if the ima ges were not taken for this purpose). Archived data from previous excavations of VIAS-University of Vienna has been chosen and the derived digital surface models and orthophotos have been examined for their usefulness for archaeological applications. The a bsolute georeferencing of the resulting surface models was performed with the manual identification of fourteen control points. In order to express the positional accuracy of the generated 3D surface models, the NSSDA guidelines were applied. Simultaneously acquired terrestrial laser scanning data – which had been processed in our standard workflow – was used to independently check the results. The vertical accuracy of the surface models generated by SfM was found to be within 0.04 m at the 95 % confidence interval, whereas several visual assessments proved a very high horizontal positional accuracy as well.

Poppy Crop Height and Capsule Volume Estimation from a Single UAS Flight

“…The 3D point cloud can be registered to a projected coordinate system by identifying and incorporating ground control points in the bundle adjustment [26,27,29,38]. A dense image matching algorithm is then employed to extract a dense 3D point cloud from the images [29,[34][35][36][39][40][41][42].…”

Mapping with strings attached: Kite aerial photography of Durai Island, Anambas Islands, Indonesia

“…In contrast to the heavy plate and film cameras of the nineteenth and early twentieth centuries, today's lightweight digital cameras make KAP a practical means of collecting large-scale aerial imagery using consumer-grade equipment. Imagery collected via KAP has been used in a variety of scientific applications, for example to investigate forest cover and stream channel characteristics (Aber, Sobieski, Distler, & Nowak, 1999); to create digital elevation models (DEMs) for studying geomorphological processes (Marzolff & Poesen, 2009;Smith, Chandler, & Rose, 2009); and to create maps of vegetation canopy structure (Dandois & Ellis, 2010), the geomorphology of periglacial features (Boike & Yoshikawa, 2003), archaeological sites (Doneus et al, 2011;Kersten & Lindstaedt, 2012;Verhoeven, Taelman, & Vermeulen, 2012), intertidal rocky shores (Bryson, Johnson-Roberson, Murphy, & Bongiorno, 2013); and coral reefs (Scoffin, 1982). Other platforms for collecting aerial photographs include balloons, poles, and unmanned aircraft systems (UASs).…”

“…aspx) and PhotoScan (http://www.agisoft.ru/). PhotoScan was selected for this project because it supports both automated and manual tie point generation, a feature not available in some programs; it does not require a priori camera location or orientation information; it handles all of the tasks necessary for orthophoto generation in a single software package; it has an active community of users who share tips and ideas through a forum (http://www.agisoft.ru/forum/) and wiki (http://www.agisoft.ru/wiki/); and models created with this software have been evaluated by researchers in several domains (e.g, Doneus et al, 2011;Gini et al, 2013;Kersten & Lindstaedt, 2012;Verhoeven, 2011).…”

Rainfall simulation and Structure-from-Motion photogrammetry for the analysis of soil water erosion in Mediterranean vineyards

“…The images acquired were processed using the commercial software Agisoft PhotoScan®, as already successfully considered in different analyses (Doneus et al, 2011;Javernick et al, 2014;Piermattei et al, 2016;Prosdocimi et al, 2015;Verhoeven et al, 2012;Woodget et al, 2015). A custom algorithm similar to the Lowe's (2004) Scale Invariant Feature Transform (SIFT) object recognition system was used by the software to determine the 3D location of matching features in multiple images.…”