Video
Heritage Preservation and Terrestrial Laser Scanning
Transcript
Karen: Good afternoon. Welcome to the NPS Archeology Program Speaker Series for Fall 2013 and Winter 2014. My name is Karen Mudar and I am an archaeologist in the Washington Archaeology Program Office.
Our speakers for today, Lori Collins and Travis Doring are co-directors of the Alliance for Integrated Spatial Technologies, which is a research center at the University of South Florida. It's working to preserve and protect the world's cultural and natural heritage through education and global engagement. Collins and Doring were early adopters of laser scanning with heritage projects using terrestrial laser scanning starting in 1999. They've worked on projects all over the world. Including Mexico, Guatemala, Greece, Spain, Egypt, and the Bahamas. Wow, I wish I had your travel itinerary.
They've partnered collaboratively with the NPS on a number of projects, most recently in the Southeast, National Capital Region, Hawaii, and the US Virgin Islands. Today, Collins and Doring are going to talk to us about some of their research. Two new technologies, such as terrestrial laser scanning, advances in imaging and photography, and 3D printing are greatly advancing our capabilities for heritage preservation and research. The ability to rapidly and accurately document the world and share that data is revolutionizing archeology and museum sciences and is creating new areas of research integration.
In this webinar, examples including the NPS collaborative projects will demonstrate the latest in heritage and archaeological documentation research and emphasize effective workflows and approaches for heritage management. Additionally, this webinar will show how these projects are bringing national parks to the classroom and changing the way we teach, learn, and interact with our past.
I see you've forwarded my presentation, so thank you very much for doing that. Terrestrial laser scanning especially as it related to documentation of historic structures and buildings is a hot topic in the NPS right now, so I'm looking forward to this talk with particular interest. Lori suggests that you ask questions during the talk rather than waiting until the end, so feel free to jump in, especially if your question involves an image on the screen. It can be really time-consuming to scroll backwards and forwards through the PowerPoint. I'll also monitor the question pane, in case it's hard for you to get to a phone.
Thanks very much for speaking to us today, Lori and Travis.
Yeah, you're welcome. For simplicity's sake, we're going to have me driving the show here, but we do have Travis as well as pretty much the full crew in here. Dr. Jeff [Duvernae 00:06:01], our PhD student Joseph Evans, and also Barb McCloud, who are all integral parts of our team. They're around for question and answer as well. I do also want to point out that if anyone's interested, if you go up to the Handout section there that I'm pointing out I think, I'm trying my pointer, here in the corner, you can actually download the PowerPoint as a PDF note file, and that will give you all of the links, because apparently this system doesn't really allow for video inclusion, so I provided links to all the external videos that we would have included in the PowerPoint.
Today's talk really is a show-and-tell, showing how we're actively doing a lot of these technologies in the park system and in particular we've chosen three projects to highlight as case studies as we go through this, but I think it will really showcase a full range of the technologies that we've been utilizing for the Park Service's benefit.
Think it has a little lag time here. There we go. Starting out with a little bit of who we are and what we do. We're actually a center at the University of South Florida that works to preserve and protect the world's cultural and natural heritage. We're in the School of Geosciences, and we actually have students from a diverse range of departments at the campus. We have mostly environmental science, geology, geography, and archaeology students, however we also employ in our center computer science and engineering students, so we get a nice mix of students coming into our center and working together on these projects. Our goal is to create an interdisciplinary atmosphere and really to impart the projects and the work that we do into the classroom.
I'll get used to this in a second. Here, this is just to give you an overview. There's our office in 3D. Our primary areas of specialization, we do a lot with GIS, we also do a lot with terrestrial LiDAR and aerial LiDAR, as well as Global Positioning Systems. A lot of cartography, geodesy activities, and as well as remote sensing and image subsurface survey work. We really bridge a lot of cross-disciplinary activities that go on between social, biological, and physical sciences. Then we work in collaborative relationships with the public, private, and national and international academic sectors.
To tell you how we're bringing TLS into these heritage applications, TLS, or Terrestrial Laser Scanning, we really don't use any one tool in particular, we're integrating a lot of different types of technologies, including things like photogrammetry, high res imagery of different types such as gigapixel and reflexive transformation imagery, which we'll talk about a little bit here. GIS and GPS and a lot of different kinds of geophysical kinds of surveys as well. Really we're doing this in an effort to be more representative if you will of what we're actually documenting. If you think of things that are recording, things like GPS where you're taking a point, a positional point, you're taking many of those positional points, with laser scanning you can be much more representational of the actual thing that you are documenting. I think you'll see that in the slides ahead.
We really have broad research themes in our different documentation projects. In particular we've benefited from a lot of these collaborative relationships with the National Park Service, where we've been focusing in on things like cultural resource management, climate change and rock art degradation, and things like NAGPRA and restoration support efforts. All of these projects have provided a great deal of information that we've been able to bring back to the classroom and allow students to be immersed in our national parks system and learn more about technology by using the national parks as a living laboratory. I should say that here at USF we also benefit from the fact that we have a high res visualization center, where we can see all of these things in 3D and we also have a rapid prototyping lab in our engineering facility where we can do a lot of 3D printing and other kinds of fabrication efforts with our 3D data.
To tell you where we are with the National Park Service, bringing this into the classroom we've got both traditional kinds of semester-long offerings, as well as continuing ed offerings, we do a lot of technology transfer with industry partners as well, continuing adult education, and we're trying to also bridge between social sciences and more traditional hard sciences, to be able to increase learning and capacity in different interdisciplinary kinds of approaches.
Today's talk will center on 3 projects that I think showcase everything that we're just talking about. Primarily we're going to talk about themes of climate change, heritage loss, and management activities. I should also point out that all of the projects that you'll see have been done in cooperation with NPS SEAC as well as our park-level cooperators, HER Dr. Margo Schwadron and funding through the CESU program which has been really important to our efforts.
The first project I'll go through is Fort Matanzas, and it's right here in our backyard in Florida. Fort Matanzas is a national landmark site that quite literally has issues of falling apart, constant repair needs. The site is made out of a really fragile resource that has a lot of imperiled kinds of mortars and binding textures to it that have to be considered for management purposes and we're losing a lot of these fragile mortar areas. The park actually is really interested in any way that they can document this in a way that will help them with management. I should also point out that our projects often have multiple applications and usages. Not only is it for management, but as you'll see, there's a big interpretive benefit and public benefit, as well as the educational benefit that comes from this.
Here we see the fort which is over, if you're familiar with Daytona in Florida, this fort is in 2003 kind of what it looks like today. This fort, though, historically has had significant problems. You can see the historic images where it's actually cracked apart in several places. What we're doing is laser scanning the present-day condition of the fort, but we're also considering how the fort changes and may be modified as a landscape through time.
To do this, we work collaboratively with the park so we're actually ahead of time getting a feel for what their needs are and what they're trying to do with the data and information as end products. For this particular project we're using midrange terrestrial laser scanning, phase-shift scanner, the faro-scanners, is what are our choices for this project. The GPS equipment we're using is various from RTK to mapping grade, using standard panoramic, spherical, and integrated scanner photography. Airborne Lidar is an important application as well in a lot of the GIS work that we do in these parks.
For the deliverables, again, in concert with the park we decide, "Okay well what are some of the things that you're going to want out of this?" For the park management aspects, standard CAD measured drawings are important. For this particular project they had HABS-HAER level drawings that were done historically, so including a lot of that legacy data in our information is also very important. Things like 3D fly-throughs, 3D sections and elevations, airborne LiDAR integration, Google Earth model production customized for them, 3D models for replicas and web interpretive aspects, as well as a final report running through all of these things. All of these things are coming from the same sets of data that we're collecting.
We do this in a number of ways. This particular project took us 2 days in the field, probably about a day and a half of fieldwork itself and about half a day of set up and logistics. We did a couple of innovative things, we used a mast system on this project where we elevated the laser scanner here to really acquire a very good coverage with the laser scanning both inside and outside of the fort structure, as well as some of the terrain aspects. Then we're using the aerial LiDAR applications to really put the fort within its terrain setting and understand the fort in terms of its landscape setting. This is a custom application and a KMZ model that was produced for the park and you'll see that we're producing things in ways that we know people can use it. Some people are GIS users, they're very comfortable with GIS, but other people - everybody can use Google Earth, so we think it's important to include our data in ways that are going to be usable for people.
Not only do we provide the park with that geodatabase, that's for their GIS users, but we're also doing things like webshare, which I'll show you a little bit more of and these custom KMZ datasets that just about everybody can use or you can keep them for select users.
This is an example of webshare. What webshare allows us to do is take our scanned projects and make them accessible either to the public or for secure sign-in. In this case, the secure sign-in allows the park to even download the scans if they want. We don't make any of the scan data public, but we are making certain derivatives of the scan data public such as still videos or still images and videos. What the webshare allows you to do is download spherical images, panoramic images in high definition. You can do full measuring within a webshare, you can move in and out choosing each one of the scanned scenes that were taken from each position including interior settings. You can navigate around the scan and actually see where the scanned position was taken from and do measurement within a plan-view map, more familiar to a GIS user. You can also embed objects within the scan, attached to the scan, maybe from that perspective, such as HABS-HAER drawings, PowerPoints that have been created, slices or other derivatives that come from the scan data and even YouTube videos.
This format that I'm showing is actually a double sort of thing going on here. I've got webshare that's actually embedded within a Google Earth. We can move in and out of formats and provide them in such a way that we get a lot of people able to use these datasets.
Where it all kind of starts is with the laser scanning. The laser scanning provides us with our model that we take from a point cloud which in this case is a colorized point cloud in this column, if you will, here. We create a solid surface polygonal mesh with our data of the fort, and we have things like texture maps that are made from external photos that have been taken of the fort that we can use to texturize the scanned data to make them more realistic. In this case, one of the points was the park was really interested in public interpretives that could be developed out of this.
I'm just going to run through this, the whole video is here, the link and you also have that in your handout. I'm going to run through some of the different views of the fort. This is from the scanned data, creating a 3D model. Because it's scan data, both inside and outside of the fort, we can actually slice through the fort and reveal what the interior setting looks like. It's the interior setting, I should say, minus the furniture and everything that's in there today. This is down to the bare walls. We can see the full surface textures, and the modeled kind of look here that's being portrayed more for interpretive and visualization kinds of efforts.
The scan data itself is all measurable and fully able to be gone back to in place of the model. The model allows us to do creative kinds of things. Here's the flag, the flag with AIST on it, so we can do a lot of different things to showcase the project as well. It's also really important for us to document the entire landscape. The landscape setting in this case is really important because it's changed so much through time. GIS and GPS applications really allow us to do that. We're bringing all of these datasets together, airborne LIDAR, old maps, legacy data, in this case we have a map of the fort location in 1867, and we can really see how in fact the landscape has changed as we move through time. The earliest here being 1742, moving on to 1867, you start to see the land form coming together here, changing in shape, here in 1971 you see a big difference in the position of the fort in addition to the landscape modification that's going on, and then in 2010.
The landscape has a big effect on what the forces such as erosion and other kinds of activities that are going on in the fort. That's why it's important for us to look at this in different ways and at different scales.
This is a CAD derivative that comes right from the scanned data so we can include all the measurable data, and we can provide this in different looks and formats for the park and it's all accurate. In this case, the laser scanner we were using is about a 2 millimeter accuracy instrument.
We're providing a lot of this content in ways, again, that a lot of people can use. Going along with the project are things like our Flickr datasets and our uploads to the website, different websites and social media applications. In this case, this is our own website where we host project pages and all of the media content, as well as the full report of the project can be found on our website. On social media we get a lot of social media attention and we also share a lot of social media applications with the Park Service. We have Facebook here, Flickr, and YouTube. All of these are getting a lot of hits and a lot of sharing going on, so we feel this is a really good way to showcase Park Service projects that are allowed to be showcased with the public. We do have some NAGPRA ones that we don't put anywhere.
The other thing that's really interesting with these projects is that we're bringing them back into the classroom. The students have gotten really engaged in the things like 3D printing or rapid prototyping. Here we have a miniaturized 3D print of Fort Matanzas that was actually done by students in my Technologies for Heritage Preservation class, and they've gotten really interested in not only the fort itself, but in landscape modeling, in classes like Dr. Doring teaches a GIS class, so we use these as projects for students to utilize in the classroom as well.
Another thing that we're doing is investigating uses for new technologies. As we move forward in time, we're thinking more and more people are going to be utilizing wearable computers such as Google Glass. Here at AIST we have a Google Glass Explorer. Joe Evans, our student here, was one of the first to be wearing Google Glass, and also now I wear Google Glass for teaching. We've been trying to incorporate this into our projects and in doing so thinking more and more people are going to have their own devices when they visit parks and they're going to be doing things like these wearable computers, so we're doing things like AR, where we're doing pattern recognition either on a place or on a tagged image.
In this case, we've actually developed poster series that are used at different events, conferences, internally and externally, where people through their own device can hook up to an aura and can actually see the videos and everything through their own device as they look at the poster. It's a way to engage the public in different ways. We can also do this at the site itself through park interpretive signs and other locations that can become enabled to provide details about the project as someone actually walks through the park. These are offering really some interesting new opportunities for using our scanned data and 3D models that come from our projects.
The next case study that we're going to look at is Hawaii. This was a hard one to do. This was another collaborative project with SEAC and this is a case study where we, again, utilized a lot of these different kinds of technologies, so we'll show you some of these, including some short range, different scale laser scanning that we did in this project as well.
This project involved two different parks that we worked in. HAVO and PUHO, and I'm not going to say the names there for shortness' sake. Here you see, these are maps of the various locations. We actually did three different locations within HAVO and one in PUHO. All of these were in an attempt to really look at using different technologies for documenting imperiled petroglyph sites.
The first example I'll show you is from PUHO. This is the Konani game boards that are so important as a feature on the landscape. How do you document these at different scales? We used an approach here. Here we're using terrestrial laser scanning at the landscape scale, documenting the entire, entire area that these features was found within. Then we're also documenting with close range scanning so that we can produce replicatable models, we can 3D print these for example, of the actual game board itself. We're able to do a scalar change here and go from really high resolution to looking out at a more complete landscape.
I think that that's really important, is that because we deal with multiple scales in our parks, we really need to think of multiple levels and multiple methods for documentation. No one scanner is probably going to fit all, if you will. We go from anywhere from longer range applications with airborne LIDAR to more landscape-level documentation strategies with our phase-based scanner here to the short range, artifact or feature documentation with more portable or shorter range kinds of laser scanning, and then also supplementing with a lot of GPS survey and GIS analyses.
Oh sure.
Lori, this is Stan Bond, how are you doing?
Hey, Stan.
I just wanted to make a comment about the papamu, having worked in Hawaii for a while. I'm reticent to just call them game boards because I think they probably provided a whole variety of functions and when we give them a name like a game board, we've given them a static meaning, but to me, I often thought of papamu as graph paper.
Yeah that's interesting. Stasack actually was noting that in his report where he was talking about the idea that maybe they conceptualized things using these, because they're so irregular, too. They're not like a checkerboard the way they've been described.
No, I mean, but they're regular in the sense that they almost always go one side is even number, like one ... Up and down might be an even and odd number, and the vertical and horizontal might be an odd number, but they're always even and odd. One number apart at least for one type of set, and then there's another set that seems to be just a series of three lines with seven pukas or seven cupules long.
Yes. It's really interesting, because I think that this technology offers us a way to explore these in a different sense, too. We've actually 3D printed a couple of these pieces and you can look at them and explore them differently, so that’s really interesting, too. Not oversimplify them, huh?
Yeah I don't like to put a functional name on them because I think they had a whole variety of functions that could deal with counting, navigation, mathematical functions, design functions, I think, along with being, using them for games. Laying out strategies, mapping. I wouldn't want to limit the idea that they were only for games.
Right, okay. Thanks, Stan. One of the other tools that we're utilizing a lot are specialized imaging techniques. Along with standard imaging, too, high res imaging. Here we see a couple of different examples of just using a little more advanced imaging. Some gigapixel imaging, where we're doing some interesting things with online gigapan images where you can zoom in and really see high resolution details and zoom out and see an entire landscape. Here we're showing, actually our NPS contributor taking photographs and actually GPS photos as well, we do a lot of, to actually, have a geo position at the same point where we're able to visualize what is there. This really helps to sort of supplement the GIS applications that can come out of these projects as well.
This is actually a feature in HAVO that is a lava tube feature that we documented using both laser scanning and some of the advanced imaging techniques. In this case, the laser scanning was really useful to provide one of these webshare opportunities that I've been talking about. Basically, the park actually will get its own secure URL and they can log into this project and they can navigate around the scanned data without having to know the full scanner software. There's no real learning curve here, you just navigate almost like you would in Google Earth, and you have full panoramic, if I could show you in a video here, but I can't, sorry. You can see this online. On YouTube we have some examples of these, on our YouTube channel. Where you can actually go around looking in panoramic image, you can switch this out and look at the additional documentation, we have videos and other kinds of things attached to this. You can move it in and out of the various scan positions. You can really move through a scan project without knowing scan software basically. There's no learning curve. It's a nice deliverable for the parks.
The other product from the same area that we did was utilizing reflectance transformation imaging. What RTI does, is essentially you have the camera in a stable position and you're taking images moving light sources. Your light sources are actually moving around at variable angles. In this case we took 60-90 stable camera shots and then we're assembling these multiple images in a software package where we're creating these polynomial texture maps, and these files can be interactively viewed and the nice thing, again, about the product that you get from this is that these files can actually be viewed in free, open-source software. There's no need to purchase software, and there's also very little learning curve. It's very easy to navigate through. This is what the interface looks like.
If you look here at this static image, this is what you see in the lava tube and this below is an enhanced specular image of the petroglyphs and you're seeing all of that enhanced detail that's able to be brought out that you don't see with the naked eye when you're in the cave itself. If you're interested in a video of this, I've posted the link within this as well.
This is just a close-up in the same site. Here's another glyphic representation of possible ancestor glyph that was found or noted in the cave by an earlier survey, so what we've done here is used RTI imaging on it to really pull out and enhance all of this detail. You can almost use it forensically to see how things were made and how much pecking and everything goes into making one of these glyphic representations.
This is in HAVO as well, different area though. Not petroglyphs this time, but actually fossilized footprints that are on the Ka’u Desert trailhead there. What's interesting about this is that there was a public interpretation, or is, rather, but it was actually an ARPA incident, back in 1994 where they had somebody actually, they believe, dig out with a shovel or other kind of implement, the footprints. This is what it looks like today. The footprints themselves are now missing. There are, however, areas out off the trail where footprints do still exist. These are extremely fragile. They can obviously be taken away or vandalized or disappear by people doing things like walking on them or not really intentionally doing anything but maybe degrading the surface as they're actually using the resource.
What we did here is actually talk with the Park Service about different ideas for bringing this site back to life, maybe through augmented reality applications or also enabling people to virtually see the 3D scanned data that we collected at the site through different, basically their own devices that they might bring into the park or have with them. The way that we documented this, this is a photograph of one of the areas that we documented. Interestingly, you can see the smudges from the person walking and we know that it was wet ash as they were moving across this, making these lasting footprints here.
We used gigapixel technology, we used a lot of advanced imaging techniques, we also documented them with GPS and we're also doing laser scanning of the landscape area all around here, and actually doing close range laser scanning on the actual footprint area itself. What we're producing from this are polygonal meshes of the footprints from the close range laser scanning is what you're seeing here. This was a white light scanner, portable white light scanner that we actually took out and used at night because it gets interfered with by light sources really easily. We did nighttime scanning at the closer range here.
You'll notice, too, in this photo maybe you'll see these little white dots. Those are actually the targets that we used for the close range scanning to be able to assemble the scans together. This white dot over here that you see is the actual target for the longer range scanner that we use, so similar principal. We're actually compiling all of these scans together into a meshed model, and surface texturing it with the images and producing things like videographics that can be shown. Here we have a video that you can go see on YouTube that actually shows how the footprints were made, walking through the desert. We actually took the same day that we created with the video with, and produced 3D print models, exact duplicates of the footprint area.
What's interesting here is that in the future we hope to work with all those to maybe take some of these replication ideas on a little bit larger scale and we can actually replace that damaged ARPA area in the interpretive center with a representation of the footprints so that people can still interact with it and see it but not have to worry about any further damages to the piece.
Similar kinds of things that I showed you for the Fort Matanzas project. Here we're giving customized KMZ files so that we can share this in ways that are easy to use, easy to access. This is actually showing some of the GPS photography that was done as well as embedding things like the YouTube videos and other kinds of content. We're doing similar things on a larger scale. This was at the Pu'u Loa site where they had 23,000+ of these petroglyph carvings there. They're very well interpreted for the public, but it's quite an effort maybe to walk out to this site, and maybe people don't understand perhaps what they're walking across or what is there.
Maybe some of this data can also be utilized for augmenting some of the interpretation that's already existing, actually bringing some of the data to the site or at offsite locations such as in the Jagger Museum or other locations at HAVO where we can provide people with 3D content as if they're at the site, however, not needing to maybe walk out to the site. Some people might have ... Things where they can't do that. We feel that this data is rich enough that it really offers a lot of different possibilities.
This is interesting here that we were able to create a digital elevation model. There's no LiDAR data. The LiDAR data from the airborne LiDAR actually cuts off right close to the coastline and so we did not have full access to aerial LiDAR for the site, and so we were able to take terrestrial scans that have been done by the HAVO staff using a longer range instrument, and we were able to create a custom digital elevation model for this site area using their data. We can really understand the terrain.
This is just an example of our GPS work where we've collected data to show the viewing platform. This is what it looks like on the ground. Park Service has made this viewing platform so that people don't have to walk across the petroglyph fields. We've created GPS hotlinked photo locations for many of these features including all of the areas that we actually scanned and did work at.
To provide the Park Service with a range of types of deliverables that they can benefit from, HAVO was really interesting, HAVO and PUHO both, because they have laser scanning as a tool that they have access to, and so it was interesting to work with them, with their scan data that they've already collected at a different scale and to actually teach and learn from one another. This was a true collaborative aspect of the project that was really beneficial.
This is an area along the trail where we're doing GPS survey. Obviously the trail itself, besides the petroglyphs, the actual ancient foot trail going into the site is very integral part of the site, and so it was really important for us to be able to document this as a landscape.
Also of real importance out here is using technologies that are non-contact and noninvasive. Everything we're doing is non-contact that I'm showing, and that's really important because people have often documented petroglyphs in ways that are actually damaging to the resource itself. This is a really good example of somebody that's done a silicone pull off of one of these petroglyph areas. You can see the actual damage that gets left behind from that kind of contact replication process. We don't need to do that anymore. Through laser scanning we can make an exact replica as you've seen with the 3D prints that are derived from the scan data.
We did just that at some of the really important features along the boardwalk that are interpretive for the public because we felt that the 3D prints would supplement things that they're doing at other offsite locations people can see and interact with these petroglyphs without necessarily having to go to the site, or they might get really excited about the petroglyphs and want to go to the site. It should have multiple benefits from that respect.
The way that we did this was using the portable white light scanner at a closer range, creating a really high definition scan of the feature itself. Here we see this is actually an image of the 3D print that was made from the scan data. This is another example. The same panel area, and again, you see we are scanning at night because we're collecting better data that way, and then here you see the 3D print that is made from that. This way we don't have to have any contact or any damages that's done to this piece in order to have this kind of rich product.
The same area being documented at the same time with GPS, with specialized photgraphy. Here we're doing night raking ... Light raking photographs at night. There you go. Really able to showcase a lot of these really faint or subtle carved areas in this way.
All of this dataset, because the landscape is so important, we're bringing this all together in a GIS. This is another deliverable package, a custom geo database where all of this data is given to the Park Service so that they could either include it in their own geo databases or they have the foundations for the beginnings of one. In this case, we've included even legacy data that's been geo-referenced, so there's been other surveys that have been done at Pu'u Loa in this case, where they've actually created mapping grids that may or may not be extremely accurate, but we've been able to geo-reference them and bring them in for consideration. Overlaying on top of them all of our GPS locations so that we can critically look at previous surveys that have been done and theories that have been developed from those kinds of surveys.
We also have complete GPS photography databases that go with this so that as you select a point with information, you get not only the XY position coordinate for this, but you also get an image of what the piece, what that area looks like. It becomes a really good informational system for the parks.
The same database can easily be converted into other formats such as KMZ in case people are not GIS users, we always like to think that people for the full range of users that might want access to this data, and not everybody knows how to use GIS. We can provide the same kind of data layers as KMZs so that you can navigate with the Google Earth or Virtual Globe platform. We can also provide geo images for GPS imaging software packages, or in this case we can lay it into a KMZ and provide all of these hotlinked or tagged image positions right available at a click of a mouse.
This is at HAVO as well, this is another area I thought was really interesting to show that we not only documented with laser scanning, but we also used the RTI imaging of this particular piece. Why I thought that this was so important is again it shows indication of damage from people doing silicone molds. Again, we don't need to have any contact to be able to both replicate and to totally understand the surface carving and be able to share it in a digital environment. Here's just a closeup showing all of the damage left from the impregnation of the silicone into the rock itself.
Through the RTI imaging we can really do things like specular enhancement, so if we've got areas of really faint carving or in this case we want to really look at those damaged areas, we can change things and alter the surface, make the surface detail much more visible by utilizing these reflectance images.
Again, the multiple ways of sharing this information. Taking all of this data and creating these 3D prints that can be handled, passed around, I use these in classes that we teach here at USF. This is a really good way to engage people and get them interested in the site and learning more.
The final project that we want to show you is a case study. Again, a really hard place to go to and do work, the Virgin Islands. Again, another petroglyph documentation, rock art documentation project as part of this climatic change investigation. Here we're working with one locale that has a dense number of carved areas that are imperiled that we documented again in multiple kinds of ways. Here we show an example of all of the GPS work, both walking into the trail. This trail was really hard to get into, the trail system itself ... You can see the terrain differences here as you come into this site, and so not everybody can visit this site. One of the other considerations we had was everything we produced from this project, we wanted to make it where it could make the site accessible in a digital environment or a virtual visitation sort of way. We felt that that was another important aspect in talking with the park on this project.
What does site look like? This is another area with these petroglyph carvings that are imperiled for a number of reasons. Atmospheric conditions, changes in surface erosion kinds of activities that are going on here. We've got people that actually come into this site and walk across this feature and continue on the trail system here. There are impacts from things like vegetation, changes in the rock surface itself that you can see evident, even in this picture. They're imperiled for a number of reasons, so how do we approach this? How do we document this in ways that we can digitally preserve this site?
Again, choosing multiple types of technologies, laser scanning at the mid-range, gigapixel, RTI imaging, and GPS, GIS, kinds of applications. This is interesting. This is the laser scan data point cloud that emerged from the project where we sliced it, creating a digital representation of the terrain through the site, so we can kind of see the site has a waterfall feature that goes from one pool and dips down to another pool. We can visualize and understand this terrain in a different way by using this scan data.
Again, creating these custom applications so that the Park Service can actually utilize this data for their everyday management, but also for things like interpretation and education. In this case, another customized KMZ database, a GIS database, as well as these webshares, are important tools for sharing and learning. This is showing you the webshare that's available for this. The webshare here has a lot of fly-through videos, taking us through the scan data, we have videos showing the RTI images of each one of the petroglyph site locations, and the user can simply navigate through each of the scan setups that we did at the site and then they can see all of the associated media that goes with that. This becomes a really one-stop shop for a Park Service manager to go through and actually visualize the whole project. They can do things like measurements, get high-def pictures, and actually interact with the scan data itself.
This is taking the scan data and actually creating a polygonal mesh product out of it. We were also interested in things like rapid prototyping, 3D printing, but also in creating interpretive videos and other kinds of content so that people could actually see and appreciate what's carved on here. The other thing that kind of came out of this, though, was our scan data picked up double the amount of known carvings. There were 13 known areas carved on this one panel face and we actually have now 26. You'll see the question marks that are by areas that have not been previously shown as containing carved elements. We are seeing things in different ways by visualizing it through the scan data applications that we're collecting.
This is one of the rock areas that has just been recently found, it was actually written about. The archaeologist at the park recently rediscovered this because Ken Wild, the archaeologist there, actually found a roll of film in his desk drawer and it was undeveloped, and when he developed it he saw that here's this carved area that nobody had previously documented. He was able to go back out to the site and re-find this piece. This is sort of what it looks like in the field. Very hard to discern what's there and you can see how it's cracking and coming apart, so very fragile and imperiled. Through things like reflectance transformation imaging and 3D scanning, you can see all of that carved detail that we've been able to extract out. Even areas that previously were not known or shown even to the trained eye. We do have an historic image of this piece where they chalked it up and even in the chalk image, they don't show any carving over to this side. We've been able to document new areas of carving.
We're actually seeing things that are hard to see. Really hard to interpret in the field, largely because we've got things like this vegetation and lichen growth that's covering it. You can't really see it or appreciate it in the field, but we can bring that out with the reflectance transformation imaging, and we can also show and share it with the 3D imaging and 3D printing.
Again, another example, this is how you see it in the field, this is how you see it with the reflectance transformation imaging, and this is how we're able to laser scan and 3D print the piece for sharing and for interpretive development. At this site, it's going to be really nice to have some of this interpretive development actually offsite at Cinnamon Bay Ken Wild wants to utilize some of these things there, because not everybody can walk out to this site, it's quite a challenge to walk out to.
Then we're doing things like creating rendered video files, video productions, that can actually fully take advantage of our 3D data and model it to include terrain and landscape kinds of activities and features like the water and the pool here, and then extract out some of the carved detail so that people can more visually appreciate what's there. If you're interested in seeing the video rendering of it, I've included the link there for you.
In conclusion, I'm hoping that you'll have some questions, we're really seeing that archaeology is changing. Technology is continuing to change and we expect lots of new things coming in the coming years, and we're trying to showcase and stay ahead of that, and showcase these things to the Park Service. We're really able to create and connect better to people in general and create better sustainable partnerships with the National Park Service through utilizing technology. We're really improving things like management, education, and outreach potential.
What's interesting is we're able to take those very same projects and showcase the National Park System in the classroom as a living lab and get students really excited and interested in the Park Service and in our nation's parks and really present a way for them to develop their own research and interest relating to what is a real world challenge or project and so they get really excited about this. This has been a wonderful opportunity for us to collaborate with the parks because it's improving the way we teach as well. That's the end. Hopefully you guys will have some questions and feedback for me, too.
Lori, thanks very much for a wonderful presentation. Do people have questions for Lori?
Did I put people to sleep?
I don't think so.
We have one here at SEAC.
Hey Lori, it's Sarah here at SEAC. I was just ... When you were talking about rebuilding the footprints at Hawaii volcanoes, I know the 3D printed material is biodegradable, so would that be a problem in making the footprints last?
No. Actually so, okay so that's interesting question because you're right, we can print in biodegradable material, but we can also print in many different types of material. In fact, we can print in metal even. In fact, there's a thing called bioprinting where people are actually printing human tissue now. We can print in all different kinds of tissue, all different kinds of material, rather, so it doesn't have to be the biodegradable kind that you're thinking about. Here in our rapid prototyping lab, we have several different kinds of printers and we print in a powder resin, we can print in a harder plastic that is not the biodegradable kind, and as I say you can print in metal and all different formats as well, so it doesn't have to be a limitation or a problem at all.
Well cool, thanks for the information.
Yeah.
Lori?
Yeah.
Hi, my name's Mike Jones. I work for the Flagstaff Area National Monument. I do GIS for them. I have some questions just related to data processing and what kind of computer processing power you need to be using to create your ... You go out in the field and you take all these points and then you develop the point clouds into something useful. What kind of computers are you using?
You know, on the low end probably 16 GB of RAM. 2 Gig processors, you need as much RAM as you can, and you need a good video card. Video cards are really important. Joe, you want to maybe chime in a little bit?
None of these processes are really GPU or IO bound, they're not intensive, so you need systems with a lot of RAM and a lot of GPU capabilities to process this data quickly.
You use desktops or servers?
In the field we're using workstations, so our preference I guess if we have to tell you a name, it would be a Dell Precision workstation for field work. They're kind of clunky and large to carry around, but they do offer you pretty much a workstation in the field, kind of processor. We're using workstations in our lab, and we do have cluster computing capability as well if we get into crazy cloud sizes. Large rendering, we have a render farm that we can push to.
Okay. Thank you.
Hi Lori, this is Karen.
Hey Karen.
It looks like a lot of the subject matter that you're working with is culturally sensitive. Do you see a need for any cautions with these types of data that you're collecting and their availability?
Definitely. This is an area, I'm actually working with Stan Bond right now on a project where we're compiling and considering a lot of ethical implications of how these technologies get utilized in parks. One of the things that's continually coming up is how much do we make publicly available and what do we make publicly available? As you can see, if you have access to a 3D point cloud, technically you can replicate it. Our choice is, in most instances, we are not making 3D data publicly available for download. We're making derivatives of 3D data available for download, but they're not anything you can actually replicate from.
I say that, then the Smithsonian goes in the complete opposite direction. They're wanting to make their whole collection available. I can sign a user agreement and I can go in and download objects and 3D print them from the Smithsonian. That's their collection, and they're doing it with pieces that are not sensitive. You definitely do have to think about this because there are implications to these datasets. You'll notice that I said at the beginning I think, that we do a lot of NAGPRA projects with SEAC and those projects are not publicly available in any format and they're rich sets of data but they're available for SEAC and the tribes and that's where that stops.
There's a lot that has to go into this, I think, from a collaborative standpoint that you really have to work with your collaborators and understand fully what can and should be made publicly available and what is sensitive. We do have to be cognizant that people are going to be able to create their own 3D data using their phones and other kinds of technologies. They already are. What are the implications of doing that with park collections and in park settings? I think we don't need to be scared of it, we just need to be aware of it. That's some of the reasons for this white paper project that we're currently working on with Stan Bond as well.
Thanks for that. Unless people out in the audience have other questions, I have another question for you.
Lori, I've got a question. This is Stan.
Hey, Stan.
In your Fort Matanzas work, I know that at Fort Matanzas there are a lot of masons marks on some of the stones. Did you record any of those as well?
Yes, we did. Actually the report on that project includes some of the still images of some of that, but we definitely have that all in 3D as well.
Great, thanks.
Yeah.
Hey, this is Brendan out in Sitka, I have a question about cost. Let's say the Fort Matanzas work, how much did that run to get all those data products together?
Fort Matanzas is probably not a good example because here's what we did with Fort Matanzas. Fort Matanzas was a small project that we took on and we did big things with it because we are actually have another contract to do the Castillo, which is the bigger fort in St. Augustine. This was a way for us to work with the Park Service, really figure out what's going to work for them, what they like best out of it, so we probably did a lot more for that project than we would typically do for the price range that we did, but I mean every project's going to be different as far as the kinds and types of deliverables that get included.
What I find interesting, I'll say that if you do these kinds of projects the way we're doing them, the Park Service definitely is getting let's call it more bang for the buck, because these are so usable in so many different types of formats. What is interesting to the way we've approached it I think, is that because these are collaborative projects, when we're done we actually turn full data sets over to the Park Service. There's no holding back on things. We're hoping that they're going to continue to develop and use these things for things like monitoring and other types of resource management into the future.
The price is going to vary depending on those deliverables. Those rendered movies and all of those, the customized geo databases and things, they do take extra time on the processing side of it. It's not our standard, it's not just registering the data, it's post-processing and doing many different things with it. These projects vary in price and scope, but I think that the three regions project that I showed which included the Hawaii and the Virgin Islands work is 100,000+ project. Primarily travel and things are included in that, too. You're getting a lot of deliverables for not a lot of cost really.
Yeah, that's fine. It's just important to know for budgeting purposes, if we wanted to put in a project.
Yeah and we'd be happy to talk with folks, too, because the idea here is that we can also work collaboratively if this is going to improve content for maybe a course or something like that, or do it a phase or in steps. There are different approaches that we can take to make sure that we are meeting budgetary needs as well. We're obviously sensitive to that, too.
Can you use laser scanning data produced by another organization?
Definitely. In fact, in the HAVO example that I used, it was interesting because they actually collected laser scan data themselves, using a longer range flight instrument for the Pu'u Loa petroglyph field, and they weren't really sure how that could be integrated. What we decided to do with that data, because it was at a different scale, it gave us a wider area terrain model capability, so we were able to bring that in and create a digital elevation model for that full area using the data that was previously collected. We're really happy to work with the Park Service. SEAC has been wonderful, opening up their data to us and making all ... Even legacy data that has absolutely no geo-reference information to it, old photographs, old drawings, things of that nature, we can incorporate into these databases in ways that are meaningful.
Okay so the specific thing that I'm thinking about is, Harper's Ferry came out and laser scanned our totem poles and the Russian Bishop's house.
Okay.
I haven't seen ... I don't know the resolution of their laser data, but I suppose it's pretty good. I'm not sure how sophisticated the products coming out of that data are going to be.
Right. Yeah, so we could utilize, for example, we could take laser scan data that's been collected as you're saying and we could create new products from it. These video renderings or any kind of 3D rapid prototyping product that you might want from that. The other thing that we can do a lot with previously collected data with is augmented reality and creating interpretive hotspot developments for actual interpretation either offsite or within your park.
Right. Okay.
Lori, this is Mike Peterson, archaeologist, Redwood National Park. Quick question here.
Hey.
I was wondering what your handheld scanner is, what model or brand that is, and also what's the resolution? Can you scan, can you pick up, detect microwears, the scratches or abrasions on lithics or chert, flakes?
Yeah, okay, I'll say that the one that we were showing on that project was one that we were actually just beta testing in the field that time. That one, it's not necessarily that I would necessarily say, "Buy this one or look at that one." That one was actually a Creaform , it was called a Ghostscan and that one it's a white light scanner, it does get really high resolution, approaching, it's sub-millimetric. The white light scanners, though, even though they have high resolution, they have a lot of limitations. If you're doing anything outdoors, you can have a lot of limitations from ambient light concerns. Shininess of objects, things of this nature.
We also, for close range scanning, we utilize a Minolta scanner as well, we utilize a fero arm scanner, so again, kind of like one size doesn't fit all, but if you're looking for a portable option, there are many out there. We also really like the F5 Mantis, we've utilized that in the field, too, for more portability. Then, also, Brinkman makes a more portable. It changes. The technology is changing so quickly and there's so many of these on the market, you can either even get some of the lower end ones that are approximating, fairly good accuracy for some of this. If you're interested, I'd be happy to e-mail correspond with you.
Okay, thank you very much.
Hey, Lori, you've talked a lot about working with partnerships and working with the parks and cooperative arrangements. Are you using the CESU program?
Yeah, all of these projects actually are CESU projects, and it's a wonderful way really for universities and the National Park Service to come together. It keeps the overhead rate down and allows us to work much more expeditiously. The other thing I really love about this program is it encourages us to share the fieldwork experience and really work together on efforts. We've been doing a lot of this, SEAC's been really innovative with taking advantage of this. We actually even have a student researcher that's housed at SEAC that helps out on some projects. There's some really interesting internships and other kinds of opportunities that we've been able to take advantage of. It's a great program.
Well thanks. Do people have questions? Lori, can you talk about any new technologies or areas of your research that you think are going to change in the next couple of years?
Yeah, I think one of the emerging areas is going to be in wearable technology such as Google Glass and other kinds of immersive, augmented reality kinds of applications. I think we're trying to see how we can take a lot of the 3D content we're creating and bring it in to these different kinds of formats. The other thing that we're really interested in is how we can utilize these to increase learning potentials for students and so we've actually been using Google Glass in the class. Another hard one to say.
We've been creating Google Hangouts and allowing students to check in to these Google hangouts and actually come with us on a project or see what we're up to. You can see through our eyes as we do things in the field. We're getting ready to do a really cool project with 96 Historic Site and SEAC in South Carolina, and we're actually going to utilize Google Glass to create a hangout that will allow people to come with us into the tunnel as we laser scan this American Revolutionary Period tunnel.
We're doing some neat and innovative things with some of these technologies that haven't quite hit the marketplace yet. Also, just with scanning in general. Cheaper, faster, smaller, and it's portable and mobile and can be mounted to things like drones and UABs and we're going to see more and more of this kind of data collection, even from our portable devices like phones, that are going to be totally accessible for the average person. Again, not to be scared of it, but the Park Service can be a leader and get out front of it and take advantage of these technologies and also know the ethical and standards limitations and applications that need to be happening. I think that there's a lot of really interesting things. It's never been more interesting to be in archaeology than in this age because I think we can do a lot of digital discovery now, and it's really improving our abilities to research and learn.
Well, this is a really great example of how the CESU program is enabling the Park Service and parks to access cutting-edge technologies.
Yeah, it's been a great way to create and generate content for the courses that we're doing as part of our curriculum and heritage preservation here at USF, too. Having these case studies has made a huge difference for student learning capacity. The students are really actively excited about these projects, like doing hands-on things that they see are having some kind of a tangible benefit, a lot of the students were creating or working on some of these 3D prints, like that Fort Matanzas piece. They get really excited when they see something tangible like that and they can relate it to a real project.
Well very good. Does anyone else have comments or questions for our speakers today? No? Well if not, I thank you both very much for speaking to us and I really appreciate your ability and cooperation to bring this very interesting talk to Park Service archaeologists.
Thanks everybody, and thanks for having us.
Description
Lori Collins; Travis Doering, 1/9/2014, ArcheoThursday
Duration
1 hour, 11 minutes, 3 seconds
Credit
NPS
Date Created
01/19/2014
Copyright and Usage Info