DOCUMENTING MACKENZIE INUIT ARCHITECTURE USING 3D LASER SCANNING PETER

5 WS DOCUMENTING A BUSINESS PURPOSE SPECIFIC DOCUMENTATION IS
A MINUTE FOR THE MEDICAL STAFF DOCUMENTING COMPLICATIONS COMORBIDITIES
DOCUMENTING ARTS + HEALTH CPD PEER BASED ARTISTS CLINIC

DOCUMENTING FARMER’S INNOVATIONS OR HOW DO PEOPLE SURVIVE THROUGH
DOCUMENTING HAZARDOUS WASTE CONTAINER INSPECTIONS ALL HAZARDOUS WASTE GENERATORS
DOCUMENTING MACKENZIE INUIT ARCHITECTURE USING 3D LASER SCANNING PETER

Laser scanning:

Documenting Mackenzie Inuit architecture using 3D Laser Scanning.

Peter C. Dawson^¹, Richard M. Levy^², Charles Arnold^³, Gerald Oetelaar^⁴, Dominic Lacroix^⁵.

Laser scanning is currently being used in various areas of the world to document ancient architecture (Ioannidis 2004; Ahmon 2004). Laser scanners record the proveniences of numerous points on an objects surface. The resulting three-dimensional images can be used to test various building scenarios, analyze activity areas in a three dimensional context, and digitally archive heritage resources threatened with destruction via erosion and industrial activities (Brizzi et al 2006; Brown 2001; Ioannidis 2004). Laser scanning may have applicability in the Mackenzie Delta region, where archaeological research has become increasingly focused on the interpretation of Mackenzie Inuit architecture, and the preservation of houses threatened by erosion (Friesen 2006; Arnold and Hart 1992). In this paper, we report on our use of 3D laser scanning to document a Mackenzie Inuit dwelling from the Pond Site (NiTs-2), located in the outer Mackenzie Delta. Specifically, we evaluate the challenges and benefits of using laser scanning in remote arctic environments, and outline potential ways that 3D computer reconstruction might shed light on the study and preservation of Mackenzie Inuit architecture.

Introduction:

The study of variability in architecture and its relationship to cultural processes has been an important subject in anthropology since the very beginnings of the discipline. This is especially true in the circumpolar world, where the preservation of architecture is often excellent. One of the first anthropologists to explore the relationship between house form and culture in this region of the world was Marcel Mauss (1906). In his book Seasonal Variation of the Eskimo, written in collaboration with Herni Beauchat, Mauss argued that variability in Inuit architecture reflected seasonal shifts in household organization. The search for cultural processes in circumpolar architecture continues to spark interest in arctic archaeology and ethnography, and many recent publications on this topic attest to this (Friesen 2004; Dawson, Levy, Gardner, and Walls, 2007; Dawson and Levy 2006a,b; Lee and Reinhardt 2006; Whitridge 1999; Patton and Savelle 2006) . The Mackenzie Delta region of the Canadian arctic, in particular, has garnered much attention (eg. Arnold and Hart 1992; Friesen 2004). Over the past several decades, researchers have attempted to link variability in semi-subterranean architecture to changes in Inuvialuit social organization through time (Friesen 2004). Unfortunately, resolving these important research questions has been hindered by the erosion of houses and sites, as well as inconsistencies in the methods employed to document and record Inuvialuit architecture, both historically, and in the present day.

Laser scanning has been used extensively in Europe and other areas of the world to rapidly and accurately document archaeological sites and heritage buildings (Guidi, Fricher, De Simone, Cioci, Spinetti, Carosso, Micoli, Russo, Grasso n.d; Monti, Fregonese, and Achille, 2003, Giuffrida, Liuzzo, Santagiti, and Andreozzi, 2005), Some of the most notiable projects involving laser scanning include the Parthenon, Herculaneum, Giza, (Stumpfel etl.al 2003, Brizza et al 2006 and Neubauer et.al 2005) . However, laser scanning has rarely been used for such purposes in Canada, where it has been confined primarily to recording small objects and heritage buildings. By way of illustration, the National Research Council of Canada, one of the first institutes to develop triangulation-based laser scanning technology in 1978, has used laser scanning and VR technology to record and model museum artifacts and paintings, most famously, Da Vinci’s Mona Lisa (Blais, Taylor, Cournoyer, Picard, Borgeat, Godin, Beraldin, Rioux, and Lahanier, 2007). The perception that laser scanning is difficult to use in remote locations, coupled with the highly technical nature of processing scanned objects, may explain why Canadian archaeologists have made little use of this technology to date.

The use of laser scanning technology in an environment as remote and challenging as the Arctic provides us with a unique case study in which to evaluate the feasibility and benefits of using laser scanning to record archaeological features in less remote areas. The Canadian arctic presents some formidable obstacles because many archaeological sites can only be accessed via helicopters or fixed wing aircraft. Laser scanning equipment is also highly sensitive to dirt, moisture, temperature, and impacts. Maintaining a steady, reliable source of power is also critical, as are adequate light levels which are necessary to obtain clear scans. Finally, the instrument needs to be operated on a stable platform that is impervious to wind.

Our successful application of laser scanning in the east channel of the Mackenzie Delta during the summer of 2007 demonstrates that these challenges can be effectively managed. Furthermore, recording architectural features in 3 dimensions and at high resolutions makes it possible to visualize architectural features in ways that are simply not possible using 2 dimensional line drawings. The resulting scans can form the basis of 3 dimensional computer reconstructions of scanned features, as well as test various construction scenarios, and interpret the domestic use of space.

We begin by providing a brief overview of the use of laser scanning in archaeology. We then discuss Mackenzie Inuit architecture, and how variability in floor layout and construction practices has been interpreted by archaeologists. Next, we describe the methodology used to record two Inuvialuit house ruins (one total, and one partial) at an archaeological site on the East Channel of the Mackenzie Delta. This is followed by a discussion of the pros and cons associated with using this approach. We conclude that short-range lasers, while offering higher levels of resolution, are not really suited to the task of recording large areas containing architecture. Rather, they are best matched to recording architectural details that may hold significant clues to constructions practices and/or social use of space. We feel that short range high resolution laser scanners, similar to the one used in this paper, could be used to supplement more traditional recording techniques (plan drawings), or used in combination with long range scanners such as the Cyrax 2500, to capture entire architectural features at varying levels of detail (Riegl, Studnicka, and llrich, 2003, El-Hakim 2005, Malinverni et al. 2003).

The Use of Laser Scanning in Archaeology

Many disciplines have benefited from the infusion of laser scanning technology. Since its inception, scanning technology has used representations of archaeological and architectural monuments as showpieces in its marketing materials. However, because of the inherent expense, it is not surprising that the majority of laser scanning applications have been in engineering, surveying and manufacturing. The expense of acquisition, staff training, and maintenance has prevented the use of this equipment from becoming commonplace in both historic preservation and archaeological fieldwork. Though the expense of this equipment represents a barrier in many fields, over the last decade many notable archaeological projects concerned with significant heritage sites have relied on laser scanning.

A great deal of early laser scanning work in archaeology and conservation focused on capturing the images of smaller object such as statues. In 1999, two research projects embarked on the scanning of Michelangelo’s statues, including the Renaissance sculptor’s famous depiction of David. One of the research groups, led by Marc Levoy of Stanford University, used a custom designed triangulation laser built by Cyberware to scan Michelangelo’s statues in Florence, Italy, including the David, the Prigioni, and the four statues in the Medici Chapel (Levoy, et.al 2000). . The scans were of a resolution high enough to reveal the artist’s chisel marks on the stone. (Koller, Turitzin, Levoy, Tarni, Croccia, Cignoni, and Scopigno, 2004). Other examples include the laser scanning of ancient cuneiform tablets by Subodh Kumar (2003) and the Mona Lisa by the National Research Council of Canada (Blais et al. 2007). While small objects allow for such levels of high resolution, it is impractical to scan buildings and other larger scale objects in this way. One way around this has been to scan extremely detailed models of ancient towns and cities, at high levels of resolution. In 2005, Gabriele Guidi and colleagues scanned the “lastico di Roma antica” a model of Rome which was built during the last century (Guidi 2003). This last project reveals a critical issue in laser scanning, which is the relationship between the size of the object, and the degree of resolution possible. The lastico di Roma antica model was both large and contained small details. This would be a situation comparable to a full sized archaeological site, where the researcher might be interested in recording an object as large as a house – a feature that is both large and contains small architectural details that might be of importance to the researcher. There is a trade off with laser scanning, in so much that efficiently scanning large areas reduces the overall resolution of the resulting image. Consequently, issues of resolution, range, accuracy, rate of capture and color depth must be evaluated when determining the type of scanner needed for a project.

Types of Scanners.

The choice comes down to three major categories of laser scanners: Pulse (time of flight), Phase (triangulation) and Modulating Light ( Boehler, and Marbs, 2002 ). Time of flight scanners can have a range up to 800m, but most work well in the 100-200 meter range. These types of laser scanners are excellent for acquiring 3D images of buildings and large sites; they have a moderate degree of accuracy for a single point (1cm-.5mm depending on the distance to the object) and have a scan rate of approximately 2000- 50,000 points/sec (Figures 1). These pulse scanners operate by measuring the time of flight required for a laser pulse to reach the surface of an object and be received by the scanner. An important advantage in this technology is their ability to operate in any lighting conditions. With the introduction of self-rotating laser-emitting heads, the speed of data acquisition has been greatly increased. With fewer setup steps and less post-processing, this technology is ideal for reconnaissance and for establishing baselines of historic sites and archaeological excavations (Finat etal 2005; Johansson 2002; Sternberg, et al 2004) .

Make/Model	Rate pts/sec	Source

Time of Flight Scanners

Leica HDS 3000	1800	http://www.leica-geosystems.com/hds/en/lgs_6506.htm
Leica Scan Station 2	50,000	http://www.leica-geosystems.com/hds/en/lgs_62189.htm
Trimble GX 30	5000	http://www.trimble.com/trimblegx.shtml
Riegl LMS-Z420i	11,000	http://www.riegl.com/terrestrial_scanners/lms-z420i_/420i_all.htm
Riegl LMS-Z390i	11,000	http://www.riegl.com/terrestrial_scanners/lms-z390_/390_all.htm
Riegl LMS-Z210ii	10,000	http://www.riegl.com/terrestrial_scanners/lms-z210ii_/210ii_all.htm

Phase and Triangulation Scanners

Lecia HDS 6000	500,000	http://www.leica-geosystems.com/hds/en/lgs_64228.htm
Faro Photon 80/20	120,000	http://www.faro.com/pdf/FARO_Photon_en.pdf
Faro LS 420	120,000	http://www.faro.com/content.aspx?ct=us&content=pro&item=5&subitem=14
Minolta Vivid 910	300,000	http://www.konicaminolta.com/instruments/products/3d/non-contact/vivid910/specifications.html

Figure 1: Selected Laser Scanners, Data Capture Rate

Triangulating laser scanners come in single and double camera versions. Triangulating scanners generally offer high spatial resolutions (less than .3 mm) with low distance ranges (an order of magnitude less than time of flight scanners) see figure 1. Capable of acquiring data at speeds greater than 100,000 points per second, these scanners offer the archaeologist the ability to acquire highly accurate 3D images of artifacts and architectural details. One issue with triangulating scanners is that, depending on the manufacturer (model and make), light levels must be within a very specific range. Too much light and the camera will not function properly. Very shiny surfaces under bright lighting will result in holes in the data set. Single camera versions work on the principal used by range finders; a known baseline distance between the mirror and camera lens allows triangulation on a point. Triangulating scanners that employ a double camera are similar to the single camera, but feature a light projector that produces a moving strip or static pattern. These patterns when viewed by the camera at a fixed distance from the light source can provide data used to determine the shape of the object. Though not capable of capturing data over a large area, they do provide accuracy in a range of .1mm to .6 mm, depending on the distance to the object and the design of the unit. Versions of this type of scanner exist that are bench mounted, making it possible automate data acquisition and making this type of unit complete to transport and setup (Dı´az-Andreu, et al 2006; Cain, and Martinez. 2004 ; Finat et al 2005; Barnett et al. 2005.)

Finally there are scanners that use a modulated light source. By varying the amplitude of the light source the camera can determine the distance from the target or object. Some of the phase scanners will split the laser beam into several components, each with a different wave length ( Boehler, and Marbs, 2002). For example, the Faro laser scanner splits light into a 76, 9.6 and 1.2 meter wave lengths. The distance to the object is determined by registration of the shorter wavelengths against the longer cycles. The range of these scanners can exceed 70m and provide resolution of .6 to 1.2mm, depending on the distance to the object. One advantage of this type of scanner is its speed. With point acquisition of over 120,000 00,000 per second and a 360-degree field of view these scanners can provide a good alternative for general survey work (Figure 1). When faced with the problem of scanning objects that are both large and contain small details, as with The lastico di Roma antica Project, the solution can be to use multiple scanners with different resolutions. Guidi et al (2003), for example, used a modulated light scanner, supplemented by a triangulation scanner, to capture the model at different levels of detail (El-Hakim et al 2005; . Malinverni , Fangi, and Gagliardini 2003; Riegl et al 2003 )

Color Acquisition.

The colour acquisition that is important in archaeology detail presents a very unique set of challenges in 3D imaging. Though it is possible to capture color data with all types of scanners, very controlled lighting conditions of the target or site is required for consistent and accurate color data to be captured. With laser scanners, color values (RGB) are recorded for each point acquired from the surface of the object. This data can be converted into texture maps that can be used to wrap the surface of a mesh, a process known as texture mapping. In creating 3d photo realistic models, digital imagery can be combined with laser scanning data. High-resolution digital cameras can be mounted directly onto a laser scanner. Using a software solution based on photogrammetry principles, it is possible to select color values from the digital image for every point in the 3D image or point cloud. In general, a digital camera can also operate under a greater range of lighting conditions ( Beraldin 2004; El-Hakim,et al 2005. Riegl,, Studnicka and Ullrich, 2003; Lambersa et.al 2007 )

Environmental Conditions.

In the lab, it is easier to have a high degree of environmental control than in the field. Clean power, lack of dust and vibration, excessive heat and cold do not concern the researcher in a lab compared to an excavation site (Neubauer et al 2005; Sternberg et al. 2004) . The laser scanning of the Herculaneum Roman Baths highlights the kinds of problems typically encountered. These relate mainly to the need for setting up multiple survey stations, the problems of differential surface reflectivity, in this case caused by moisture on the walls (Brizzi et.al 2006) (Reference). Consequently, when selecting a laser scanner for fieldwork, issues of transport, reliability, and outside operating conditions must be considered. All scanners have operating limits. Ordinarily, scanners will not work in dusty, wet or excessively hot or cold environments. Though some scanners have been designed to minimize the impact of dust entering a unit through the use of heat sinks rather than fans, most units are sensitive to heat above 100⁰ F and below freezing 0⁰C. (check). Operating in the rain is probably not advised nor recommended for any scanner. Though future scanners may not require a fixed platform, today, most scanners need a stable and fixed platform from which to operate (Neubauer et.al 2005; Blais, Picard and Godin 2004 ) NCR current research project). Given that scanners are must have a line of sight view to the target (they must be able to see the object), placing a scanner on scaffolding that shakes under the use of the operator will result in unusable data.

For anyone who operates a scanner, transportation to the site becomes a serious issue. There are only a few scanners whose size and weight allow them to fit on board in the overhead compartment or under the seat of an airplane . The use of G-force Ccases, though reliable, will probably not guarantee that falling off the back end of a truck or an airport baggage conveyor does not destroy or damage the unit. In addition, ancillary equipment must be transported including, targets, digital camera, computer laptop, connecting cables, uninterrupted power supplies (UPS), generators, tripods, tarps and tents (Sternberg et al 2004) (Figure 2, Site 2008). Being self-sufficient in the field may require the transport of several hundred pounds of equipment, easily exceeding the normal luggage allowance with commercial airlines.

DOCUMENTING MACKENZIE INUIT ARCHITECTURE USING 3D LASER SCANNING PETER

Figure 2, Laser scanning with the Minolta Vivid 910, House Site 3, summer 2008 – In this photograph some of the equipment needed for the project is in view in cluding the scanner and tripod, USP tarps, Gforce case.

Budgetary Considerations.

Though price may be a limiting factor in the purchase of a scanner, the type of data needed for the project should guide the actual choice. In reality, one scanner may not be sufficient to achieve the research objectives of a project. Two scanners may be required: one for the general survey of the site and another for capturing detail including, relief, architectural ornamentation and artifacts detail (El-Hakim, et al 2005; Malinverni et al 2003; Riegl, Studnicka, and Ullrich, 2003 ) . Ultimately, the expense of laser scanner acquisition may be so great that purchase is prohibitive. A better solution may be to contract with a firm that will charge several thousand dollars a day for data collection. Though costly, contracting out may be more cost effective, particularly when working under contract or when the need to capture data is infrequent in nature. As many first generation scanners are approaching 10 years in age, a used market may emerge that will make acquiring equipment for research and teaching more affordable. Much of the older equipment captures data at the same resolution as newer ones, but at a lower speed, which in a teaching or research environment is not as critical as in business.

The Mackenzie Inuit Sod House.

1. Ethnographic Descriptions of Inuvialuit Architecture.

During the 19^th century, the MacKenzie Delta region was home to one of the most populous groups of Inuit in the circumpolar world. Among the most identifiable visible markers of these societies are large semi-subterranean, cruciform-shaped houses used during the winter months. The architecture of these enigmatic dwellings has been summarized extensively in the ethnographic literature by Franklin (1969), Nuligak (1966), , Petitot (1876), Stephanson (1914), Richardson (1828), Stringer (nd), and Whitakker (1937). Out of these observations emerges a dwelling “type” with a fairly standardized set of architectural features. These house forms are described as having three alcoves or sleeping platforms, opening off of a central room designated as a main chamber. A fourth alcove, or extension, forms part of the entrance passage. Ethnographic accounts describe the main chamber as constructed from four corner posts outlining a square, with dimensions ranging from 8 to 12 feet on a side. The corner posts consist of inverted tree trunks set into the ground with their roots serving as crotches for four stout logs, which form the main ridgepoles of the structure. Roof height inside these dwellings was recorded as 6 feet, with the roof being constructed from split logs with the flat sides facing inward. The ceilings and walls of the alcoves consisted of split logs resting obliquely against the ridge poles forming the four sides of the main chamber, not unlike a lean to. According to Whittaker (1932), the lower ends of the logs “were set on the earth about two feet beyond the square, and leaned against the upper logs, until the spaces were filled” (check reference from Friesen 2004). The structure would have then been covered with sod blocks for the purpose of insulation (Dawson and Levy 2006, Levy, Dawson and. Arnold 2004; Levy . and Dawson, 2004 ). Figure 3

DOCUMENTING MACKENZIE INUIT ARCHITECTURE USING 3D LASER SCANNING PETER

Figure 3 Reconstruction of an Inuit Sod House, Property of the Author

Inside the dwelling, a skylight, which also served as a chimney or alternate entrance, provided some ambient light (Friesen 2004:229). These were typically set into the roof, usually above the katak or door, and covered with ice or semi-transparent skin (Stefnsson 1914:160; Petitot 1876:15). With the exception of the window, the only source of heat or light mentioned in the ethnographic literature is the stone lamp which, along with the body heat of the occupants, maintained an ambient temperature of 5º to 15º centigrade. The number and placement of the lamps varies from author to author. According to Petitot (1876:14-15), the number of lamps depends on the number of families present in the dwelling, each family having its own source of heat. They are supported on a double row of stakes as close as possible to the floor of the main chamber near the base of the closest support post. A grill located above the lamp provided a space for heating or thawing items or for roasting meat. The traditional cruciform houses at Shingle Point on the Beaufort sea coast apparently had five stone lamps, one near each of the main posts and a fifth at the back of the alcove opposite the door. Nuligak (1966:16) notes that three of these lamps or kudliks, usually one in each corner, burned all the time, while more lamps were lit when needed.

The alcoves served as sitting and sleeping places for family members, as well as work spaces and storage areas, and were elevated from six inches to two feet above the floor of the main chamber. Although dimensions vary between authors, the alcoves are typically depicted as trapezoidal extensions from the two corner posts of the main chamber (see Richardson 1826:216 as an example). The floors of the alcoves, which sometimes had a gentle inclination forward, were constructed of split logs with the flat surface facing up. Stefansson (1914) observed that these boards were of irregular lengths and rarely met up with the walls of the dwelling. A long, narrow, slightly curved entrance passage, partially excavated into the earth and partly covered with blocks of ice provided access to the winter house. This entrance passage, which was 15 to 20 feet long by two and one half feet high leads of a trap door in the slanted floor which is covered by a piece of fur Stefansson 1914:159-160). Whittaker (1937) observed that “the best of such houses, built on a hillside, would have the entry through a long passage, leading to a trap door in the floor (get actual quote from Whittaker from Friesen 2004:228).

2. Variability in House Design.

On the surface, ethnographic descriptions present Inuvialuit architecture as relatively homogeneous, emphasizing the cruciform house over all other variations. However, Friesen draws attention to Petitot (1876) , Whitakker (1937), Stringer (n.d), and Richardson 1828), who all seem to acknowledge that while three alcove houses were the norm, single and double platform dwellings were also constructed. Interestingly, it is in the archaeological record that significant variability begins to emerge. In a recent summary of variability in Inuvialuit architecture, Friesen (2006) notes that while cruciform shaped houses dominate at the major East Channel sites of Kuukpak and Kitigaaryuit, those found to the west on the Yukon North Slope tended to have two alcoves. Friesen (2006) also observes greater variability further to the east at Nuvugaq and in the Anderson river area, where sites appear to contain a mix of two and possibly three alcoves. Mackenzie Delta architecture also varies through time. Early Thule houses are characterized by single alcoves located at the rear of the dwelling, and an external kitchen. These dwelling styles are progressively replaced by larger two alcove houses lacking external kitchens (Friesen 2006:182). While the use of external kitchens is completely unknown in pre-contact dwellings, they were occasionally depicted by ethnographers during the late 19^th century. These features are usually represented as conical tents attached to the front of the tunnel (Friesen 2006:182). Friesen (2006) has suggested that the erection of these kitchens above ground, or in the entrance tunnel, may explain their absence in the archaeological record.

3. The Relationship between House Form and Culture.

The architectural variability described above have been linked to demographic changes experienced in the Delta through time, as well as attempts to competitively emulate societies that emerged at major East Channel sites such as Kuukpak and Kitigaaryuit (Friesen 2006). The efficient exploitation of beluga whales, and participation in lucrative trade with the HBC post situated at Fort McPherson, contributed to the rise of powerful and ambitious lineage heads at these East Channel Sites (Friesen 2006:183-184). Groups in other regions such as the Anderson River area and Nuvugag, may have tried to imitate their more powerful neighbours in a number of different ways, including the construction of large and distinctively shaped houses (Friesen 2006:184).

Researchers such Friesen (2006) have argued that important social processes may be reflected in the synchronic and diachronic variability apparent in Inuvialuit architecture. Exploring these ideas in greater detail requires fine grained architectural data. Laser scanning may offer a partial solution, in that provides a rapid and highly accurate means of recording architecture at levels of detail not possible with two dimensional drawing. Freehand drawings, mapping architectural features within grids, theodolites, and total stations have all been used to record Mackenzie Delta house ruins at archaeological sites. Rather than capture a small number of reference points, typically one or two shots per object, laser scanning captures millions of points. These point clouds can then be used as the basis for computer reconstructions to test hypotheses about construction practices, the organization of domestic space, and the use of architecture as symbols for prestige and social advantage (see Dawson and Levy, 2006; Levy, Dawson, and Arnold, 2004).

Is laser scanning a feasible in the Canadian Arctic, however, given the logistical and environmental challenges, and scanning objects as large as semi-subterranean house ruins? In order to evaluate the suitability of laser scanning for such purposes, we employed this technique to record architectural data from an important Inuvialuit site on the East Channel of the Mackenzie river.

Description of Site Location: The Pond Site

The Pond site is on the near site of the pond shown in the photograph. The pond drains through extensive foreshore flats into Kugmallit Bay. NiTs-2 is located on the west shore of Richards Island at 69^o20.6’N and 134^o03.3’W. It is adjacent to, and takes its name from, a creek-fed pond that flows into Kugmallit Bay near the mouth of the East Channel of the Mackenzie River. Several clusters of shallow depressions are visible at the site, and bone can be seen eroding from the banks of the pond. Evidence from archaeology and Inuvialuit oral histories informs us that the shores of Kugmallit Bay were occupied since about 1300 AD by several regional groups of ancestral Inuvialuit. The Pond Site is approximately x km south of Kuukpak, which is remembered in oral histories as the main village of one of these regional groups, the Kuukpangmiut.

The Prince of Wales Northern Heritage Centre carried out excavations at Kuukpak over the course of several field seasons in the 1980s. The remains of several semi-subterranean driftwood and sod houses (igluyuaruit in Inuvialuktun) were excavated, ranging in age from approximately 500 to 300 years before present. With one exception, these houses had alcoves with raised sleeping platforms along three sides, and a long tunnel entering into the dwelling at the fourth side. As mentioned previously, these types of cruciform shaped houses were the common type of winter dwelling when Europeans first ventured into the area. The Prince of Wales Northern Heritage Centre conducted excavations at two of the house depressions at the Pond Site in 1989. The excavations revealed that both were the remains of fairly substantial driftwood and sod houses, but unlike the Kuupuk cruciform houses, the Pond structures appeared to have had two sleeping alcoves.

Radiocarbon dates showed that the structures excavated in 1989 were approximately 600 years old, or about a century before the dated houses at Kuukpak. An interpretation advanced at the time postulated that the Pond Site was abandoned sometime after 600 years ago, when a build up of the foreshore flats that today separate the site from Kugmallit Bay interfered with hunting beluga whales from this area. It was also postulated that the people who abandoned this area moved downstream and established their winter houses at Kuukpak.

In 2002 the Prince of Wales Northern Heritage Centre returned briefly to the Pond site in order to test whether ground penetrating radar was effective in detecting subsurface archaeological features in permafrost environments. House 3 was similar to other surface features at the Pond and Kuukpak sites that had proven, upon excavation, to contain substantial architectural remains and was chosen for the study. Although the GPR readings were somewhat ambiguous, they did suggest that there sufficient buried deposits to be detected using GPR technology. Plans to return to the Pond Site in 2003 to excavate House 3 in order to test the GPR results were postponed until 2007, when the opportunity to carry out further research was provided by funding through the BOREAS program. During the summer of 2007, two structures were excavated at the Pond Site. The first of these, House 3, was completely excavated. The second dwelling (house 4) was only partially excavated, due to lack of time (4).

DOCUMENTING MACKENZIE INUIT ARCHITECTURE USING 3D LASER SCANNING PETER

Figure 4, House Site 3 (right) and Site 4 (left)

METHODOLOGY

1. Laser Scanner Selection: Impact on Operation

During the summer of 2007, a single high resolution, triangulating scanner (Minolta Vivid 910) was used to record the excavation of two Inuit Sod houses located in the Mackenzie Delta, N.W.T. Ideally, two scanners would have been chosen; one to quickly scan the area multiple times, the other to only scan detail as it merged from the site. If two scanners had been selected, decisions would have had to be made on site concerning what was significant enough to dictate the use of the higher resolution scanner. One advantage of the Vivid 910 is that it guaranteed that the data captured of the entire excavation would be at a high resolution (0.3mm) (Figure 2). The advantage of having high-level detail of the entire site is that the significance of a particular area does not have to be made on site, thus avoiding the problem of later discovering that data is missing for an area of interest. Having a single scanner with significantly high resolution avoided this issue. The Minolta Vivid 910 is good for close range work and scans fairly quickly. It is possible to scan an area approximately 30 cm by 30 cm at a distance of 2.4 meters from the unit in a few seconds. However, given the field of view for the Minolta is between 1.2 and 2.4 m depending on the choice of lens, additional time should be reserved for scanning. Because significant overlap is needed for registration of images, a meter square can require up to 25 scans. Furthermore, issues of occlusion may require considerably more scans in order to acquire faces of objects not visible from a single vantage point. Each scan can constitute several hundred thousand points and can be stored as a text file from 3-6 MB in size . An area of a meter square can take several hours for data capture. This includes moving and setup up the equipment, refocusing, and movement of any tarps needed to shield the area from direct sun light.

Capturing color always presents significant issues during data acquisition. Under direct sunlight detail can be lost, especially when the materials are highly reflective. One solution is to use controlled artificial lighting. In a lab environment this can be accomplished fairly easily. In the field, shielding the site from direct light requires the use of tarps, tents or temporary structures. Lighting the area with daylight fluorescent will improve data capture, but will require a more elaborate setting than is sometime possible to establish in the field. One alternative is to set up a camera on a bipod that can be used to photograph a high-resolution digital image from overhead. Registration of the digital image with the 3D data can occur during post-processing. To assist in this registration, 3D targets should be used on the site. Small spheres that can be placed on a grid can greatly assist in the registration of each scan using software designed to align and optimize 3D data sets.

2. Logistics of Laser Scanning

In this study of two Inuit Sod Houses located in the Mackenzie Delta, a 1-meter grid was established over each site. Using a Leica (Brand, model) total station, critical points were measured from an established datum within the site to an accuracy of .1 mm (check what was your limit). A Nikon (model) with a XX ( Peter, what kind of camera was used in this project?) mm lens mounted on a bipod (Model Make – 2 or four legs? ) was used to create a photographic montage of the site. As part of the logistical planning for this project, all equipment was moved by commercial airlines and helicopter to the site. The equipment included a laptop computer, the Minolta VIVID 910 laser scanner (wt 25 lbs) in its protected Pelican box (Model no.1550, weight 13.5 lbs.), a Manfreto (model) tripod (weight approx 25 lbs), a daylight florescent light, tarps, targets, generator and two uninterrupted power supplies (UPS). The advantage of the UPS units is that they provide clean reliable power for the scanner. While one USP is powering the scanner, the generator can recharge the other, thereby offering continuous scanning during day. In addition to providing power to the scanner and laptop, a UPS can be used to light the portable daylight fluorescent fixtures. It is recommended that larger units (rated above 1200 va) be used for this type of work, providing power for several hours without any recharging.

Scanning in the Field

Over 8 hours (Dominic how many hours did you spend on each site? ) was needed for scanning each site. For House 3 and House 4, data was acquired from three locations (figure 2) . Even with good coverage achieved by scanning from multiple positions, some holes in the data were inevitable. To minimize potential data loss, several pie shaped scan spaces were created from each scanner position. Rotating the scanner about 20-25 degrees created a new set of scans that later need to be assembled into a single scan space.

Before the actual data is captured, the unit requires to be focused. Small adjustments in the focal range have a significant impact on data capture. A change as small as 10mm can result in a scan with holes. Operators should be prepared to spend a significant portion of time on this basic operation of focus. Scanning bright shiny objects can be problematic. Hot spots can result in a complete loss of data or a hole. Soil, especially dry soil, can be extremely reflective. Even soils that are kept damp by spraying water may be too reflective to achieve good results. During the course of this project, the sky was only dark for a few hours. Translucent plastic tarps were used to help shield the site from excessive glare from direct sunlight. A doubled over blue plastic tarp reduced the light to within operating levels needed by the Minolta.

Temperature and rain could have presented considerable problems. The Minolta VIVID 910 has an operating temperature of 10⁰ to 40⁰ C and should not be used when the relative humidity is greater than 65% (Konica Minolta webpage 2008: http://www.konicaminolta.com/instruments/products/3d/non-contact/vivid910/specifications.html ).

Fortunately during the course of the excavation, the weather did not create any serious problems. The temperature range was not below freezing nor was there any significant period of rain during the course of the excavation.

The data digital data is stored on 512 mb compact flash cards. Each card can store approximately 150 scans which can be download later to a card reader or to a PCMCIA card on a PC for backup and analysis. During the course of the project approximately 600 scans were saved on these compact flash cards. These scans were then checked in the field on a PC loaded with Polyworks10 (http://www.innovmetric.com/Manufacturing/what_overview.asp) , an application designed for assembling and processing the 3D images. It is always advised to check data for its integrity in the field as soon as possible. If holes in the data are found due to issues of occlusion or lighting, additional scans can be completed. The window of opportunity available to rescan a site is generally of very short duration for several reasons:

Excavations are conducted one strata at a time, once soil is removed they are lost forever.
After an excavation is completed, the area is sometimes filled to preserve detail for future and can no longer be scanned unless re-excavated.
Over a brief period of time, the state of the site can change. Slumping ofmaterial or settlement will result in changes in surface features.

Processing 3D Images

Acquiring data in the field is the first step in creating a 3D virtual image of a site. With PC technology it is now possible to assemble large 3D data sets into a single registered image. However with large data sets, there are still some limitations on the ultimate size of the point cloud that practically can be assembled on today’s PC. Working with the Minolta VIVID 910 scanner over the course of several days produced over one hundred million90, 000,000 points for both sites. (check no of points). The process for assembly of the data requires that each scan be registered or positioned to within the tolerance of the point accuracy of the scanner. The process is relatively simple in principle, if ultimately time consuming. In our research Polyworks Version 10.0 was used on a Dell Precision 650, Pentium with dual Pentium IV 3.05 Ghz CPU’s and a NVIDIA Quadro FX 3000 G card. After the first scan is brought into the workspace, each subsequent scan is opened and registered to the base scan using known targets in both scans as a control points. It is helpful if overlap is sufficient to virtually see the same three targets in both scans. When this is not possible, applications like Polyworks can register a set of images by identifying the same points in each scan. For example the end of a stick or small rock was used to more tightly match the two images. Usually, a minimum maximum of three points will be needed in both images to begin the matching process. Polyworks can merge each image within one standard deviation of point accuracy (check). Once this step is completed, subsequent images can be registered by repeating this process . Polyworks eliminates points in the overlap region, creating a more efficient 3D representation of the site (Figure 5).

With large data sets it is possible to exceed the computational capability of a PC. One strategy is to assembly sections of the site. Once each section is completed and optimized they can brought together as a single model of the site. Ultimately, the question of purpose must be considered in processing a 3D image. Researchers may need as accurate a model as possible for taking measurements. Most imaging processing applications will intelligently remove points where they are not needed, and still maintain the integrity of the object. However, for archaeological data with its high level of detail throughout the site, this strategy may not sufficiently reduce the number of points before the object or site begins to loose important features. One approach is to use applications that allow point clouds to be re-sampled on the fly. This gives access to high-resolution point clouds within the resolution of the screen display without forfeiting detail. High performance video cards designed for computer aided design (CAD), will permit the greatest access to the these larger data sets.

Once the point cloud is registered, a mesh can be created for all or part of a site. This surface can consist of a TIN (triangulated irregular network) of connected points. In Polyworks, as in most image processing applications, this mesh was optimized and reduced in complexity for viewing the data set in 3D. One limiting factor in creating optimized mesh files for the two study sites was the number of features located within the site. These features include small pebbles, rocks, and the texture of the wood, with its open grain and cracks. Preserving all the features sets a lower boundary on how many polyfaces can be removed during the process of optimization. Another factor in optimization was the number of small holes that existed in the assembled data set. This can result in non-continuous mesh files increasing the complexity of the form. In this work it was possible to produce a file for House 4 that consisted of approximately 200,000 polyfaces that could serve as an armature for texture mapping (Figure 63). Texture mapping can add a realistic impression of the site. Using digital images from a the Nikon SLR taken in normal daylight offered a more realistic model when compared with the color data from the Laser scanner. With a blue tarp used to shield light from the site, color data capture with the Minolta had a bluish-purple hue (Figure 5).

DOCUMENTING MACKENZIE INUIT ARCHITECTURE USING 3D LASER SCANNING PETER

igure 5: Site 4, Point cloud (left), Point cloud with vertex colors (right).

Figure 6: Polygonal Model with draped digital imagery

A variety of multimedia products were created from the model. A navigable world was created using Virtools, an application used for creating games and simulations. This virtual world allows the user to explore the site in virtual space. For display on the web and with older computers a series of QTVR movies were also created of the site. The advantage of QTVR movies is that they can be viewed on most computers. Even machines that are ten years old can view these QTVR’s. Virtual panoramas and virtual objects can be viewed with a free download player from Adobe. Finally, it is possible to share the original 3D data set with other researchers, which perhaps makes having laser scanning an important addition to the archaeologist’s toolkit. The ability to revisit the detail of the entire site at .3 mm accuracy makes laser scanning worthwhile, even given the challenges of this undertaking.

Lessons Learned and Future Directions.

This research illustrates that laser scanning can be used to capture 3D data at remote sites. Though issues of logistics will always present challenges, laser scanners are able to withstand the jolts and shaking of travel when placed in G-Force cases. Once on site, it is possible with a UPS and generators to operate in the arctic the during summer months. Having two scanners, one for general site description and one for scanning artifacts and details would have made data collection and processing quicker. However, it was possible with a single higher resolution scanner to capture a detailed 3D image of an archaeological excavation. In conducting this research, several lessons were learned. Having only recently acquired the Minolta Vivid 910 prior to its use in the field, greater experimentation with the use of targets would have simplified the assembly of the final data set. In our work we used glass shot for slingshots placed on a grid of approximately 20 cm. Unfortunately, the shot were too reflectively leaving a hot hole in the data. In the future, these small spheres will be coated with a non-reflective paint. Also if the spheres had been painted several different colors, their identification during data processing would have been made easier. Furthermore, given the need to reduce the direct light over the entire site, a portable garage or shelter could easily be transported to the site. These portable structures come in sizes from 12 to 16 feet wide and lengths of 20 feet are common. Assembled from small sections they are ideal for transport by helicopter. Made of metal or plastic tubing covered with a plastic coated fabric they can be staked to the ground. Commercial version of this type of shelter that can survive extreme conditions can create an enclosed space where equipment is safe from dust, rain and snow. This would have made it be possible to create the necessary controlled lighting conditions, making data capture data optimal. Using an array of daylight flourescent lights would result in fewer shadows and hot spots and better color control over the site being subject to changes in daylight during the course of a day. Though adding at least 4 –6 hours in setup, these portable shelters would probably be a great time saver and would guarantee better data capture.

Having established the potential of using laser scanning for recording architectural remains at remote arctic sites, the next obvious question is what benefits does it bring? As mentioned previously, variability in Inuvialuit architecture is an important research issue as it likely reflects social changes in Mackenzie Delta societies. As with many new applications of technology, the “need” usually follows its “acquisition”. In other words, laser scanning at high resolutions may reveal more variability in Inuvialuit architecture than has previously been recognized. The excavation of House 3 indicates that it is not typical of other structures at either the Pond Site or the nearby Kuukpak sites. With the exception of parts of the tunnel entrance, for example, driftwood structural elements were scarce. The structure also appeared to have only one room; there was no evidence of raised sleeping platform(s). Based on evidence from the excavations, it appears that this structure was constructed by digging into the ground approximately xx cm for the floor of the main room, and approximately xx for the floor of the entrance tunnel. Sod blocks appear to have been arranged around the edge of that part of the depression that formed the main room, forming walls of undetermined height. There were no floorboards, and in the absence of evidence of a wooden roof it is likely that this dwelling had a skin superstructure, presumably supported by small diameter logs, some of which were found in the excavations. Once processing is completed, our high resolution image of this unusual dwelling may shed further light on how all of these architectural elements functioned together to create the intact dwelling. Our scans may also reveal further variability that might have gone unnoticed during excavation, and was therefore not recorded in field notes or in plan drawings. Herein lies one of the most useful aspects of laser scanning. By capturing the excavation as a high resolution, 3 dimensional image, the researcher can return to the image repeatedly, and examine architectural elements exactly as they were during the excavation. Capturing architectural data at these levels of resolution also means that they can become the basis of true to scale 3D computer reconstructions of Mackenzie Inuit house forms. Such a model has already been constructed, but was built as a composite of various archaeological and ethnographic information on house form and construction (see Dawson and Levy 2006, Levy, Dawson, and Arnold 2004). Future plans are to use laser scanning data to reconstruct an actual house at the Pond Site. Such a model could be used to test various building scenarios, simulate site formation processes associated with the abandonment and collapse of these house types, and even examine how environmental various such as the distribution of light and shadow might have influenced the organization of domestic space (for an example of this, see Dawson, Levy, Gardner, and Walls, 2007). [1]

The expense of laser scanning, coupled with the technical expertise required to process the images once captured, mean that this approach is likely not suited for every project. However, technology is a moving target, and as the technology advances, costs will likely come down, and image processing will become more automated. We would therefore encourage our colleagues to explore potential applications of this technology to their own research programs.

REFERENCES (these have been cited above)

Barnett, T., A. Chalmers, M. Díaz-Andreu, G. Ellis, P. Longhurst, K. Sharpe & I. Trinks (2005) '3D Laser Scanning For Recording and Monitoring Rock Art Erosion', International Newsletter on Rock Art (INORA) 41: 25-29

Beraldin, J.A. 2004 Integration of Laser Scanning and Close-Range Photogrammetry - The Last Decade and Beyond, published in the XXth International Society for Photogrammetry and Remote Sensing (ISPRS) Congress. Commission VII, pp. 972-983.

Istanbul, Turkey. July 12-23, 2004. NRC 46567. – (Sources of Error.)

Blais, F., Picard, M., Godin, G. 2004 Accurate 3D Acquisition of Freely Moving Objects, Second International Symposium on 3D Data Processing, Visualization and Transmission. Thessaloniki, Greece. September 6-9, 2004. NRC Publication Number: NRC 47141.

Boehler W, M.Bordas, A. Marbs 2003c Investigating laser scanner accuracy. Proc. CIPA XIXth Int. Symposium, 30 Sept. -4 Oct., Antalya, Turkey. Pp 696-702. (see figure 8 for error for time of flight scanners)

Boehler, Wolfgang and Andreas Marbs, 2002 3D Scanning Instruments, CIPA, Heritage Documentation - International Workshop on Scanning for Cultural Heritage Recording _ Corfu, Greece

M. Brizzi, S. Court, A. d’Andrea, A. Lastra and D. Sepio 2006 The Experiences of the Herculaneum Conservation Project, The 7th International Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST (2006) np.

Cain, Kevin, and Philippe Martinez. 2004. Multiple Realities: Video Projection in the Tomb of Ramsses IIin Mindscapes, VR and Cultural Landscapes, Maurizio Forte (Editor). pp 1-14

Dı´az-Andreu, Margarita, Christopher Brooke, Michael Rainsbury, and Nick Rosser. 2006. The spiral that vanished: the application of non-contact recording techniques to an elusive rock art motif at Castlerigg stone circle in Cumbria, Journal of Archaeological Science 33: 1580-1587.

P. Dawson, and R. Levy, 2006 Using 3D Computer Models of Inuit Architecture as Visualization Tools in Archaeological Interpretation: Two Case Studies from the Canadian Arctic, Field Archaeology, 2006, 185-195.

Doneusa, M, W. Neubauerb, 2005 A Institut für Ur- und Frühgeschichte, Franz-Kleingasse, 3D Laser Scanners on Archaeological Excavations, CIPA 2005 XX International Symposium, 26 September – 01 October, 2005, Torino, Italy np ( Issues of oblique scanning angles and earth and dirt)

El-Hakim, S.F., J. A. Beraldin, L. Gonzo, E. Whiting, M. Jemtrud, VA Valzano, 2005 Hierarchical 3D Reconstruction Approach for Documenting Complex Heritage Sites, The XX CIPA International Symposium. September 26 –October 1, 2005. Torino, Italy. NRC 48229.

Finat J., and M.A.Iglesia Santamaría , J.Martínez Rubio, J.J.Fernández Martín, L.Giuntini, J.I.San José Alonso, A.Tapias, 2005 The Roman Theatre of Clvnia: Hybrid Strategies for Applying Virtual Reality on Laser Scanning 3D Files, Proceedings of the ISPRS Working Group V/4 Workshop 3D-ARCH 2005, Mestre-Venice, Italy, 22-24 August, 2005.

Javier Finat, J. J. Fernndez-Martin, J. Martnez, L. Fuentes, and J. I. Sanjose. 2005. Some experiences in 3D laser scanning for assisting restoration and evaluating damage in Cultural Heritage, pp. 543-551 in Lasers in the Conservation of Artworks LACONA VI Proceedings, Vienna, Austria, Sept. 21-25, 2005.

Johansson, M. 2002 Explorations into the Behavior of Three Different High-Resolution Ground-Based Laser Canners in the Built Environment, Proc. CIPA WG Int. Workshop on scanning for cultural heritage recording.

Lambersa, Karsten and Henri Eisenbeiss, Martin Sauerbier, Denise Kupferschmidt, Thomas Gaisecker, Soheil Sotoodeh and Thomas Hanusch 2007 Combining photogrammetry and laser scanning for the recording and modelling of the Late Intermediate Period site of Pinchango Alto, Palpa, Peru, Journal of Archaeological Science, Volume 34, Issue 10, October 2007, Pages 1702-1712.

Levy, R. and P. Dawson and C. Arnold, Reconstructing Traditional Inuit House Forms Using 3D Interactive Computer Modeling, Journal of Visual Studies, 2004, 19(1):26-35.

Levy, R. and Dawson, P, From Laser Scanning to Virtual Worlds: The Reconstruction of Traditional Inuit Houses, New Media Research Network Conference, Proceedings, 2004.

Levoy, Marc, Kari Pulli, Brian Curless, Szymon Rusinkiewicz, David Koller, Lucas Pereira, Matt Ginzton, Sean Anderson, James Davis, Jeremy Ginsberg, Jonathan Shade and Duane Fulk. 2000. The Digital Michelangelo Project: 3D Scanning of Large Statues, Proceedings of ACM SIGGRAPH 2000. pp 131-144, July 2000.

Malinverni E.S, G. Fangi, G. Gagliardini, 2003 Multi-Resolution 3D Model by Laser Data, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XXXIV, Part 5/W12. DARDUS – Università Politecnica delle Marche, Via Brecce Bianche, Ancona Italy

Minolta Konica, VIVID 910 Specifications 2008

http://www.konicaminolta.com/instruments/products/3d/non-contact/vivid910/specifications.html

Neubauer, W., M. Doneus, N. Studnicka, J. Riegl. 2005. Combined high resolution laser scanning and photogrammetrical documentation of the pyramids at Giza. Proceedings of the CIPA 2005 XX International Symposium, 26 September – 01 October, 2005, Torino, Italy.

Polyworks10, http://www.innovmetric.com/Manufacturing/what_overview.aspx

Riegl, Johannes, M. Studnicka, and A. Ullrich, 2003 Merging and Processing of Laser scan data an high-resolution digital images acquiring with a hybrid 3D laser sensor – Riegl LMS-Z and digital imagery, CIP Antalya. *also published in

cipa.icomos.org/fileadmin/papers/antalya/136.pdf

Sternberg, S. Kersten, Th, Jahn, I, Kinzel, R. 2004, Terrerstrial 3D laser Scanning – Data Acquisition and object Modelling for Industrial As-Build Documentation and Architectural Applications, ISPRS TH-17 Laser Scanning Acquisition and Modeling Techniques, XX ISPRS Congress, Istanbul, Turkey.

Stumpfel, Jessi, Chris Tchou, Tim Hawkins, Philippe Martinez, Brian Emerson, Marc Brownlow, Andrew Jones, Nathan Yun, and Paul Debevec. 2003. Digital Reunification of the Parthenon and its Sculptures, pp. 41-50 in Proceedings of the 4th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage VAST (2003), D. Arnold, A. Chalmers, F. Niccolucci (Editors) © The Eurographics Association 2003.

LIST OF FIGURES

Figure 1: Selected Laser Scanners, Data Capture Rate

Figure 3 Reconstruction of an Inuit Sod House, Property of the Author

Figure 4, House Site 3 (right) and Site 4 (left)

Figure 5: Site 4, Point cloud (left), Point cloud with vertex colors (right).

Figure 6: Polygonal Model with draped digital imagery

Figure 1. Types of scanners: Pulse, Triangulation, Phase.

Figure 2. Houses 3 and 4.

Figure 3 House 4, textured mapped model – model contains approximately 200,000 polyfaces.

REFERENCES (this is the list of all references. Not all were used in the paper)

General Works, Overview (Found in DB)

[1] Godin, G, JA. Beraldin, J. Taylor , L. Cournoyer, M. Rioux, S. El-Hakim, R, Baribeau F. Blais, P. Boulanger, J. Domey, M. Picard, 2002 Optical 3D Imaging for Heritage Applications, IEEE Computer Graphics & Applications 22 (2002) 24.

Johansson, M. 2002 Explorations into the Behavior of Three Different High-Resolution Ground-Based Laser Canners in the Built Environment, Proc. CIPA WG Int. Workshop on scanning for cultural heritage recording. http://www.isprs.org/commission5/workshop/

(comparision of time flight scanners. Cyrax 2500, Optech ILRIS-3D, Riegl LMS Z210

Mabs, Andrea 2004 Experiences with Laser Scanning At i3mainz i3mainz, Institute for Spatial Information and Surveying Technology, FH Mainz, University of Applied Sciences, Holzstrasse 36, 55116 Mainz, Germany,[email protected]

Istanbul, Turkey. July 12-23, 2004. NRC 46567. – (Sources of Error.)

Boehler, Wolfgang and Andreas Marbs, 2002 3D Scanning Instruments, CIPA, Heritage Documentation - International Workshop on Scanning for Cultural Heritage Recording _ Corfu, Greece

Standards

Metric Survey Specifications for English Heritage, English Heritage National Monuments Record Centre, 2003 Great Western Village, Kemble Drive, Swindow, England. 2003 (reprinted from 2000 edition )

Details on Laser scanning see:

Mills, Jon and Barber, David 2003 AN ADDENDUM TO THE METRIC SURVEY SPECIFICATIONS FOR ENGLISH HERITAGE - THE COLLECTION AND ARCHIVING OF POINT CLOUD DATA OBTAINED BY TERRESTRIAL LASER SCANNING OR OTHER METHODS Version – 11/12/2003 11:01 AM ISBN 1 873592 547 This document was prepared for English Heritage by the School of Civil Engineering and Geosciences, University of Newcastle upon Tyne. It was supported by English Heritage’s Archaeology Commission under grant number 3378 MAIN.

David Barber,D and Mills. J. and Bryan,P 2003 TOWARDS A STANDARD SPECIFICATION FOR TERRESTRIAL LASER SCANNING OF CULTURAL HERITAGE School of Civil Engineering and Geosciences, University of Newcastle upon Tyne, Newcastle upon Tyne 2. English Heritage Metric Survey Team, English Heritage, York.

Archaeological Excavations

Merging Data with Differrent Resolutions

cipa.icomos.org/fileadmin/papers/antalya/136.pdf

Site Excavation – Cyrax 2400

Ristevski, John, Protzen, Jean-Pierre and Addison, A., 200 Environment Creation From Laser Scan Data at Tambo Colorado, Center for Design Visualization, University of California, Berkeley, USA 2 Department of Architecture, University of California, Berkeley, USA.

El-Hakim, S.F., J. Fryer, and M. Picard 2004 Modeling and Visualization of Aboriginal Rock Art in the Baiame Cave, Proceedings of ISPRS XXth Congress. Istanbul, Turkey. July 12-23 2004. pp. 990-995. NRC 48048.

Beraldin, J.-A., M. Picard, S. El-Hakim, G. Godin, E. Paquet, S. Peters, M. Rioux, V. Valzano, V.Bandiera, 2005 A Combining 3D Technologies for Cultural Heritage Interpretation and Entertainment, SPIE: Electronic Imaging Videometrics IX. San Jose, California, USA. January 16-20, 2005., NRC Publication Number: NRC 47404.

Cyrax

Optech, Laser scanning of a Roman Theatre

Finat J., and M.A.Iglesia Santamaría , J.Martínez Rubio, J.J.Fernández Martín, L.Giuntini, J.I.San José Alonso, A.Tapias, 2005 The Roman Theatre of Clvnia: Hybrid Strategies for Applying Virtual Reality on Laser Scanning 3D Files, Proceedings of the ISPRS Working Group V/4 Workshop 3D-ARCH 2005, Mestre-Venice, Italy, 22-24 August, 2005.

Reigl

This paper describes the 3D modelling of Pinchango Alto, Peru, based on a combination of image and range data. Digital photogrammetry and laser scanning allow archaeological sites to be recorded efficiently and in detail even under unfavourable conditions. In 2004 we documented Pinchango Alto, a typical site of the hitherto poorly studied Late Intermediate Period on the south coast of Peru, with the aim of conducting spatial archaeological analyses at different scales. The combined use of a mini helicopter and a terrestrial laser scanner, both equipped with a camera, allowed a fast yet accurate recording of the site and its stone architecture. In this paper we describe the research background, the 3D modelling based on different image and range data sets, and the resulting products that will serve as a basis for archaeological analysis.

Rieko Kadobayashi, Ryo Furukawa, Yukiko Kawai, Daisuke Kanjo, Jun N. Yoshimoto, 2003 Integrated Presentation System for 3D Models and Image Database for Byzantine Ruins, Proceedings of the ISPRS Workshop on Vision Techniques for Digital Architectural and Archaeological Archives (ISPRS XXXIV-5/W12), pp. 187-192 (July 2003).

Case Studies:

ROCK ART

Herculaneum

Parthenon:

GIZA

Cases where the Minolta was used in Scanning Historic Sites

Dı´az-Andreu, Margarita, Christopher Brooke, Michael Rainsbury, and Nick Rosser. 2006.. Journal of Archaeological Science 33: 1580-1587. (The spiral that vanished: the application of non-contact recording techniques to an elusive rock art motif at Castlerigg stone circle in Cumbria)

Cain, Kevin, and Philippe Martinez. 2004. Multiple Realities: Video Projection in the Tomb of Ramsses IIin Mindscapes, VR and Cultural Landscapes, Maurizio Forte (Editor). pp 1-14

Ellis, Gavin. 2004. A conversion pipeline: From laser-scanned data to high fidelity rendering, 6 pg. Proceedings of CESCG 2004, Vienna, Austria.

Bruck, Joanna, 2005 New Light on the earliest Neolitic In the Dee Valley, Aberdeenshire, PAST No. 50, July 2005. p 1-6. Not listed on her website.

Deljkic, Edin, Jana Jovicic, Nadir Badnjevic, and Senad Biser 2006 Virtual Bosnia and Herzegovina National Museum, 7 pg. Proceedings of CESCG, Vienna, Austria.

Barnett, T., A. Chalmers, M. Díaz-Andreu, G. Ellis, P. Longhurst, K. Sharpe and I. Trinks. 2005 3D laser scanning for recording and monitoring rock art erosion. International Newsletter on Rock Art (INORA) 41: 25-29.

Mensi GS 200

Balis, Vaios, Spyros Karamitsos, Ioannis Kotsis, Christos Liapakis and Nikos Simpas. 2004 3D - laser scanning: Integration of point cloud and CCD camera video data for the production of high resolution and precision RGB textured models: Archaeological monuments surveying application in Ancient Ilida. WSA2 Modelling and Visualization, FIG Working 2004, Athens, Greece, May 22-27, 2004.

Shape Grabber

Farouk, Mohamed, Ibrahim El-Rifai, Shady El-Tayar, Hisham El-Shishiny, Mohamed Hosny, Mohamed El-Rayes, Jose Gomes, Frank Giordano, Holly Rushmeier, Fausto Bernardini, and Karen Magerlein. 2003 Scanning and processing 3D objects for web display in Fourth International Conference on 3-D Digital Imaging and Modeling, 2003, pp 310-317. 3DIM 2003.

Multi Media and Historic Resource Management

Addison, Alonzo, Doug Macleod, Gerald Margolis, Beit Hashoah, Michael Naimark, and Hans-Peter Schwartz. 1995 Museums without walls: New media for new museums. SIGGRAPH 1995: 480-481.

Alonzo C. Addison. 2001 Virtual heritage: Technology in the service of culture. Proceedings of the 2001 conference on virtual reality, archaeology, and cultural heritage, pp 343-354.

P. Dawson, R. Levy, D. Gardner, and M. Walls. Simulating the Behavior of Light Inside Arctic Dwelling: Implications for Assessing the Role of Visual Perception in Task Performance, World Archaeology, 39(1) 17-35.

R. Levy, and P. Dawson, 3D Imagining as a Tool in the Computer Reconstruction of a Thule Whalebone House, IEEE, Multi-Media Spring 2006, 78-83.

P. Dawson, and R. Levy, Using 3D Computer Models of Inuit Architecture as Visualization Tools in Archaeological Interpretation: Two Case Studies from the Canadian Arctic, Field Archaeology, 2006, 185-195.

P. Dawson, R. Levy, 2005 Explore the Ideological Dimensions of Thule Whalebone Architecture in Arctic Canada, Internet Archaeology, 18:1.

Levy, R. 2005 The Role of Laser Scanning in the Preservation and Renovation of Historic Architecture, 3D Digital Imagining and Modeling – Applications of Heritage, Industry, Medicine and Land, Padova, Proceedings.

Guarnieri A, R. Levy, 2005 A. Vettore, Integrated Modeling Systems for 3D Vision, 3D Digital Imagining and Modeling – Applications of Heritage, Industry, Medicine and Land, Padova, Proceedings..

Levy R. and P. Dawson, 2005 From Laser Scanning to Virtual Reality: The Art and Science of Archaeological Reconstructions, Ed Media, World Conference on Educational Multimedia, Hypermedia & Telecommunications, Proceedings..

Levy, R. and P. Dawson and C. Arnold, Reconstructing Traditional Inuit House Forms Using 3D Interactive Computer Modeling, Journal of Visual Studies, 2004, 19(1):26-35.

Levy, R. and Dawson, P, From Laser Scanning to Virtual Worlds: The Reconstruction of Traditional Inuit Houses, New Media Research Network Conference, Proceedings, 2004.

Notes:

Following procedures worked out through previous excavations at Pond and other nearby sites, a 2m x 2m grid was established over House 3, positioned in a manner that would coincide with the presumed layout of the house. As excavations proceeded, it was apparent that House 3 was not typical of other structures at the Pond and Kuukpak sites:

• With the exception of parts of the tunnel entrance, driftwood structural elements were scarce
• The structure appeared to have only one room; there was no evidence of raised sleeping platform(s)
• Artifacts, animal bones and other cultural remains were scarce when compared to other excavated structures

Based on evidence from the excavations, it appears that this structure was constructed by digging into the ground approximately xx cm for the floor of the main room, and approximately xx for the floor of the entrance tunnel. Sod blocks appear to have been arranged around the edge of that part of the depression that formed the main room, forming walls of undetermined height. There were no floorboards, and in the absence of evidence of a wooden roof it is likely that this dwelling had a skin superstructure, presumably supported by small diameter logs, some of which were found in the excavations.

In many parts of the Arctic where people lived in dwellings that had superstructures made from driftwood or whalebone in the cold season, and in skin tents in warm seasons, a type of structure called a qarmat was sometimes built for the fall season. There are many variants of these types of structures, but they typically are semi-subterranean, have an entrance tunnel, and are roofed with skins. House 3 appears to have all of these attributes. Furthermore, the scant nature of the cultural remains suggests that it was occupied only briefly. Although qarmat have not been detected at Kuukpak, there appears to be one of these dwellings at the nearby Cache Point site, which was occupied a century or so before the dates indicated by dates obtained for the Pond site. If House 3 indeed was a qarmat, this raises the question of whether it was used contemporaneously with the more substantial winter dwellings but during a different season, or whether it relates to a different event. Radiocarbon dates, analysis of faunal remains and artifacts should help answer this question, but for now we have a clue in the form of a harpoon head recovered from the floor of House 3.

References from Endnotes

[1] M. DÃaz-Andreu, C. Brooke, M. Rainsbury, N. Rosser, The spiral that vanished: the application of non-contact recording techniques to an elusive rock art motif at Castlerigg stone circle in Cumbria, Journal of Archaeological Science 33 (2006) 1580-1587.

Figure 1. Scanners Rates of Data Capture

Make/Model Rate pts/sec Source

Time of Flight

Leica HDS 3000 1800 http://www.leica-geosystems.com/hds/en/lgs_6506.htm

Leica Scan Station 2 50,000 http://www.leica-geosystems.com/hds/en/lgs_62189.htm

Trimble GX 30 5000 http://www.trimble.com/trimblegx.shtml

Riegl LMS-Z420i 11,000 http://www.riegl.com/terrestrial_scanners/lms-z420i_/420i_all.htm

Riegl LMS-Z390i 11,000 http://www.riegl.com/terrestrial_scanners/lms-z390_/390_all.htm

Riegl LMS-Z210ii 10,000 http://www.riegl.com/terrestrial_scanners/lms-z210ii_/210ii_all.htm

Phase Scanners

Lecia HDS 6000 500,000 http://www.leica-geosystems.com/hds/en/lgs_64228.htm

Faro Photon 80/20 120,000 http://www.faro.com/pdf/FARO_Photon_en.pdf

Faro LS 420 120,000 http://www.faro.com/content.aspx?ct=us&content=pro&item=5&subitem=14

Minolta Vivid 910 300,000

http://www.konicaminolta.com/instruments/products/3d/non-contact/vivid910/specifications.html

1 Department of Archaeology, University of Calgary, 2500 University Dr. NW. Calgary, AB. T2N 1N4

2 Faculty of Environmental Design, University of Calgary, 2500 University Dr. NW. Calgary, AB. T2N 1N4

3 Prince of Wales Northern Heritage Center, Yellowknife, NWT.

4 Department of Archaeology, University of Calgary, 2500 University Dr. NW. Calgary, Alberta.

5 Department of Archaeology, University of Calgary.

DOCUMENTING PRESENCE THE AUTHOR THE BODY AND THE NATION
DOCUMENTING THE INSTITUTIONAL DIALOGUE ON ASSESSMENT OF STUDENT LEARNING
GUIDELINES FOR USING AND DOCUMENTING COURSE GRADES AS A

Tags: architecture using, whalebone architecture, inuit, architecture, mackenzie, laser, using, scanning, documenting, peter