3D modeling of archaeological objects: Advantages, limitations, and applications
Digital 3D modeling and data analysis has had a huge impact on the field of anthropology in recent decades. One major impetus has been computed tomography (CT) technology, which produces two- and three-dimensional images of an object with computer-processed x-rays. Because these images can show both the interior and exterior architecture of an object, CT scanners can be extremely useful for non-destructive analyses of fossils and ancient remains, as well as living organisms and modern materials. Furthermore, with recent advances in micro-CT technology for high-resolution tomographic imaging, such analyses can be done on a microscopic level. The increasing use of this technology is reflected in the acquisition of CT and micro-CT scanners by museums, universities, and research centers worldwide.
Perhaps even more than CT imaging, 3D surface scanning has become widely used in archaeology for a variety of purposes. This technology can create 3D models with laser or structured light, offering many benefits of 3D analysis at a fraction of the cost of a CT scanner. While they do not have penetrative x-ray technology that reveals the internal structure of an object or the relative density of its components, surface scanners can digitally render the external surface with multiple lasers or stripe patterns and full color (RGB) photo data. When NextEngine’s 3D scanners were made commercially available in the mid-2000s, these low-cost desktop scanners became one of the first and, consequently, one of the most widely used 3D scanners by archaeologists around the world. Convenient to carry and operable with a laptop, these instruments offered researchers the advantages of portability and affordability in recording objects in collections, repositories, and laboratories as well as in field settings. Today there are a variety of manufacturers of low to high-cost 3D scanners (e.g., Breuckmann, David, and Artec), providing a variety of options in the particular instrument(s) best suited to specific needs and interests.
When selecting a 3D scanner for archaeological objects, there are a number of common criteria among users in different subfields and areas of interest of archaeology. Resolution is a top priority in general, and most surface scanners can achieve a maximum resolution that ranges from several dozen to several hundred microns (1). Speed of data acquisition is another factor, especially when it is necessary to scan objects that are numerous and/or complex or only accessible for a short period of time (for scanning in particular or study in general). Although data capture speeds vary from hundreds to thousands of points per second, scanning time is also dependent on scanning protocols and the type of system used: while stationary systems can acquire scan data with an automated platform that rotates the object up to 360 degrees, enabling users to perform other tasks simultaneously, handheld systems can scan objects of any size and provide greater control over the data capture process. Finally, the completeness of data capture is a matter of concern for most archaeologists, whose typical objects of interest may have reflective, light-colored, or transparent exterior surface material (e.g., glass, enamel, or metal) that requires additional steps or adjustments in scanning protocols depending on the type of scanner used.
As a physical anthropologist, I am interested in the advantages, limitations, and applications of 3D modeling for human skeletal remains – particularly potential uses for quantitative analysis such as geometric morphometrics. After we acquired a NextEngine laser scanner in our laboratory at the University of California in Santa Barbara (UCSB) in 2007, my colleagues and I performed several validation studies on 3D model-based measurements of the human skull. We assessed the repeatability and precision of cranial volume and surface area measurements using 3D models created by different scanner operators using different protocols for collecting and processing data, showing intra-observer measurement errors of 0.2% and inter-observer errors of 2% of the total area and volume values, respectively (2). Our results suggested that observer-related errors did not pose major obstacles for sharing, combining, or comparing such measurements. In comparing the errors of the cranial coordinate data of landmarks of different types recorded with a Microscribe 3D digitizer and from 3D laser scanner models, we found that these measurements had overall standard deviations of ±0.79 and ±1.05 mm, respectively (3). However, the 3D digitizer yielded the most precise coordinate data for landmarks defined primarily by biological criteria (Type I landmarks), while the 3D laser scanner models yielded the most precise coordinate data for landmarks defined primarily by geometric criteria (Type III landmarks). These findings indicated that the suitability of certain landmark types as reference points for geometric morphometrics depends on the method by which they are measured. Furthermore, crania displaying poor preservation and surface discoloration yielded larger measurement errors, especially for the 3D model measurements (Fig. 1).
Archaeological artifacts are no less amenable than human remains to 3D modeling, with applications that range from enhanced surface visualization (6) to digital repair and physical replication (7). Although protocols may differ in how the artifact is prepared for scanning (e.g., powder treatment) or scan data are collected (e.g., without color information), the resulting 3D models are subject to the same potential limitations and benefits for quantitative analyses. In some cases, analytical methods can be applied to both cultural and biological objects with minor modifications but entirely different objectives. For example, we found that our contour-based method of analyzing human crania with elliptic Fourier analysis (Fig. 2) could be extended to lithic projectile points (8). Obviously these artifacts do not have any anatomical landmarks, unlike crania where such features can be important for orienting and comparing morphology between specimens. However, it was possible to define and use landmarks based on object orientation to define geometric planes for the 2D contours, with digital tools in reverse engineering software (Fig. 3).
Once these data were obtained, the mathematical and statistical operations of the analysis were very similar, showing variation in bifacial asymmetry and flake scar patterns that may relate to differences in knapping skill and technique in Clovis point manufacture.
At the “ArcheObjects3D” workshop hosted by Archéorient at the University of Lyon on June 29-30, 2016, the above issues and many others were presented and discussed. The meeting was a valuable opportunity for an international and cross-disciplinary group of researchers to share experiences and insights about 3D modeling for advancing research in archaeology. Beyond the technical and analytical difficulties that can arise with 3D model-based approaches to archaeological objects, there are additional considerations and potential challenges in the management, preservation, dissemination, and downstream costs of 3D images and datasets. However, with open communication and scholarly collaboration, it seems very likely that this technology will continue to broaden and enrich archaeological research far into the future.
- Slizewski A., Semal P. 2009. Experiences with low and high cost 3D surface scanner. Quartär 56: 131-138.
- Sholts S. B., Wärmländer S. K., Flores L. M., Miller K. W., Walker P. L. 2010. Variation in the measurement of cranial volume and surface area using 3D laser scanning technology. Journal of Forensic Sciences 55: 871-876.
- Sholts S. B., Flores L., Walker P., Wärmländer S. 2011. Comparison of coordinate measurement precision of different landmark types on human crania using a 3D laser scanner and a 3D digitiser: implications for applications of digital morphometrics. International Journal of Osteoarchaeology 21: 535-543.
- Shearer B. M., Sholts S. B., Garvin H. M., Wärmländer S. K. 2012. Sexual dimorphism in human browridge volume measured from 3D models of dry crania: A new digital morphometrics approach. Forensic Science International 222, 400.e401–400.e405.
- Sholts S. B., Walker P. L., Kuzminsky S. C., Miller K. W., Wärmländer S. K. 2011. Identification of group affinity from cross‐sectional contours of the human midfacial skeleton using digital morphometrics and 3D laser scanning technology. Journal of Forensic Sciences 56, 333-338.
- Neiß M., Sholts S. B., Wärmländer S. K. 2014. New applications of 3D modeling in artefact analysis: three case studies of Viking Age brooches. Archaeological and Anthropological Sciences (june 2014): 1-12.
- Hollinger R. E. et al. 2013. Tlingit-Smithsonian Collaborations with 3D Digitization of Cultural Objects. Museum Anthropology Review 7: 201-253.
- Sholts S. B., Stanford D. J., Flores L. M., Wärmländer S. K. 2012. Flake scar patterns of Clovis points analyzed with a new digital morphometrics approach: evidence for direct transmission of technological knowledge across early North America. Journal of Archaeological Science 39: 3018-3026.
Sabrina Sholts is a curator of physical anthropology at the Smithsonian’s National Museum of Natural History (NMNH) and the director of the Smithsonian Institution Bio-Imaging Research (SIBIR) Center. Her research focuses on environmental health, infectious disease, and skeletal variation in past human populations.