Last summer, master’s student Eli Crane found himself in total darkness in a cave in the jungle, surrounded by six-inch cave crickets and bats, with his feet stuck in mud. That’s not the sort of place most people think of when they think of electrical engineering, but that’s just where he wound up, testing a unique new type of scanning device.
After helping develop a new, portable 3-D scanner with his adviser, Dr. Laurence Hassebrook, Crane mentioned the device to his friend Dr. Chris Begley. Begley, an archeologist and professor at Transylvania University, suggested Crane accompany him – along with the scanner – on an upcoming trip to Honduras to document pre-Columbian artifacts near the Rio Platano. That’s how Crane came to find himself in a cave staring at crickets as big as his palm.
“We hiked for an hour or two to get to the cave,” Crane says. “It was probably the coolest thing in my opinion.” Begley suggested the location because it was renowned for holding many artifacts. “It had a creek running through it, two or three feet deep. We had to wade in the creek to get into the cave, so we floated the Pelican case through,” Crane says
That Pelican case held about sixty pounds worth of technology: inside was a prototype form of the Rotate and Hold and Scan (RAHAS) scanner. “It is a radically new concept in three-dimensional scanning,” says Hassebrook, a professor in the department of electrical and computer engineering. “The device operates on the processors in conventional digital cameras and power efficient pattern projection systems. There is no laptop necessary.”
The RAHAS prototype consists of a structured light emitter topped by two off-the-shelf digital cameras, one running in still mode and one running in video mode. The projector shines a pattern of bright, parallel white lines onto an object, while the cameras capture both the surface appearance of the object and the way the bright lines interact with the object’s surface. Based on how the lines fall on the surface, computer software can calculate the texture geometry of the object and obtain usable 3-D information from what is essentially a series of still, 2-D images.
The process is non-destructive and, perhaps more importantly, can be done onsite. Many of the pre-Columbian artifacts in the Honduran rain forest are “petroglyphs,” or lines carved onto rocks (usually boulders resting near a stream). Because of erosion, it can be difficult to see the patterns accurately. However, because of the large size of the rocks, transporting them back to a lab for analysis is prohibitive. By capturing 3-D data of the petroglyphs with the RAHAS prototype, archeologists can analyze them in depth, perhaps for the first time.
When Crane took the RAHAS prototype into the cave, most of the remaining artifacts had been moved to one area. “Mostly they were bowls, pots or grinding stones,” Crane says. “Chris [Begley] would point to some and tell me it was marble, that one was something else – but they all looked the same to me, all brown and rough.”
Working by the light of his headlamp, Crane opened the wet Pelican case and began to set up the equipment, a job that typically took him at least 15 minutes. “Because it was a cave we didn’t have to worry about the light being too bright, so we could scan freely,” he says.
Next, Crane laid the empty, hulking Pelican case back on the ground to use as a platform for scanning objects. On top of the case he laid the calibration grid, an L-shaped piece of white foam board with a pattern of nine black circles, arranged like a tic-tac-toe board on each side. In the lab, the RAHAS prototype would be placed a precise distance from the object being scanned, but conditions in the field meant that Crane had to make the best guess he could and rely on the calibration grid to determine the object’s exact distance and orientation later in the lab when he got back to Kentucky.
Crane then set the object to be scanned on top of the case and switched on the RAHAS prototype, throwing bright lines across the surface of the object. After recording the first series of images, he reached up and manually rotated the light pattern 90 degrees over 10 seconds. In the end, the actual scanning took 20 seconds, while setting up took nearly half an hour. Crane also had to keep notes. “I had a notebook, and I wrote down every scan, where, when, what I took a picture of and in what order, so that we could organize and tabulate data later,” he says. And all this had to be done by the light of his headlamp.
“It didn’t look like a jungle initially,” Crane says. When he first arrived in Honduras, “it looked kind of like Kentucky.” He traveled with Begley, who has 15 years experience in the jungles of Central America’s “Mosquito Coast,” as the region is called, and Jorge Salaverri, a renowned naturalist and guide who holds an M.A. in forestry from West Virginia University. Begley’s research has been featured in documentaries on the Discovery Channel, The Travel Channel and the BBC. Salaverri, a native of the Mosquito Coast region, has organized educational ecotours for 15 years and has organized logistics for a number of documentary film projects by the BBC and others. They also traveled with several other local guides.
At one point, “we basically had to crawl up tree branches half submerged in the water,” Crane says. Salaverri told the group that he had often seen fer-de-lance, a native poisonous pit viper, nearby, so the team needed to be very careful.
Crane did see a single snake, but luckily it was too far away. “Most snakes in the jungle are poisonous,” he says, “but they won’t kill you right away.” Salaverri told him that it usually takes five days or so for the venom to work, so there’s usually time to get out of the jungle. “He said you’ll be pretty uncomfortable until then, though,” Crane says.
The petroglyphs are on rocks in and near the Rio Platano, several days journey from where they initially set out. The party traveled by boat up the river, stopping occasionally to document artifacts they found and set up camp for the night. “Chris would point at things lying on the ground. They all looked the same, just broken fragments of rock, but he would just point at something and say, ‘Scan those.’”
The jungle was humid and hot, even at night. “And at night, it was dark,” Crane says. “In the jungle you have this multilayered canopy over you between all those trees – so at night, it can get pretty dark.” Because the RAHAS prototype needs to be able to project bright light on the surface, scanning at night was easier. During the day Begley and Crane improvised a covering for the equipment using a black tarp supported by some bamboo rods to block out light.
When the group finally reached the petroglyphs, they found an island in the middle of the Rio Platano where they made camp. The petroglyphs in the area were on large boulders strewn around the river – some in the middle of six-foot water, others near the shore. That evening, Crane and Begley made notes about the petroglyphs’ locations and worked with Salaverri and their guides to set up a schedule to scan the rocks during the night.
Why are the petroglyphs there in the first place? It’s a mystery, and the reasons proposed are all speculation. They are usually thought to be an early form of symbolic communication, or “pre-writing.” Many of the petroglyphs in the Rio Platano area are geometric patterns or pictures of animals, such as monkeys or snakes. Crane says Begley told him that one of the most elaborate designs – showing a crocodile – is perhaps a map of the surrounding area. Whatever their significance, Begley and Crane documented around two dozen petroglyphs with the RAHAS prototype. The cultural or religious significance of the artifacts will need to be determined later.
Crane is using the analysis of the petroglyphs as his master’s thesis. On returning to Kentucky he downloaded the scans onto his computer, and is working to reconstruct the 3-D data. Because the RAHAS scanner they used was a prototype, work has been slow going, but Crane says that he’s working on cleaning up the data. Hassebrook has been working on the RAHAS scanner as well; recently he’s developed a miniaturized, motorized version that enables the user to take a laptop computer out into the field along with it.
Begley may take the improved scanner on upcoming trips to Suriname, Guyana and Costa Rica, where field conditions will be similar to those in Honduras. There’s no word on whether Crane will be accompanying him again – for now, he’s happy to be back in the lab, where it’s air conditioned, and he knows he doesn’t need to worry about bats, crickets or snakes.