Ecologists don’t always need expensive and cumbersome equipment to measure vegetation in the wild. Scientists at Rice University have found that the modern heads-up display works quite well.
Rice researchers created aMicrosoft HoloLens as a mixed reality sensor to power VegSense, their application for measuring understory vegetation, the plants that grow between the forest canopy and the floor.
A proof-of-concept study conducted by graduate student Daniel Garczynski and biologist Lydia Badreau shows that VegSense can be a viable alternative to traditional, low-cost field measurements.
Their study in Methods in Ecology and Evolution shows that a combination of hardware and software does an excellent job of quantifying relative mature trees in the wild, which is one indicator of overall forest health.
Gorczynski came up with the idea to try HoloLens, which is typically marketed as a productivity tool in manufacturing, healthcare and education. He developed open-source software for the device and noted that while the combination is less effective at picking up saplings and small twigs, there is plenty of room for improvement.
Gorczynski said he was introduced to mixed reality sensing while studying at Vanderbilt University and recognized its potential for biological research. “It seemed like a natural fit,” he said. Gorczynski pitched the idea to Baudreau in 2019 shortly after his arrival at Rice.
The combination of off-the-shelf hardware and custom software costs much less than the lidar-based systems (for “light and range detection”) most commonly used in 3D field research, said Gorczynski, who developed VegSense on a platform aimed more at on 3D games and interactive experiences than hard science.
Field tests at Houston’s Memorial Park have shown that, at least for mature trees, a smaller solution is just as good. In his case study, VegSense easily identified 48 of 50 such trees in the target area, a circle about 30 feet in diameter that Garczynski walked around, looking up, down and around to create a 3D database. (“Imagine a star with a circle around it,” he said, describing the data collection scheme.)
“For this study, we wanted to be really intentional about trying to replicate more traditional measurements of understory structure,” Garczynski said. “We tried to get that level of detail.”
What he sees when he scans the environment is a holographic grid pattern that tracks the surfaces of the vegetation. “What’s really cool about it is that you can see what the scanner is picking up and also the places you missed,” Garczynski said. “The idea is to get the net to cover as much vegetation as possible, because that’s what gives you the best scan.”
“The results were so good that Dan quickly wrote them up for publication,” Badreau said, noting that Garczynski expanded his testing of the gear on a subsequent field trip to Tanzania, which is home to one of 15 rainforests in a recent rainforest study by the Rice group.
“This device can contribute to a lot of great environmental research, especially because it’s so cost-effective,” she said. “Collecting vegetation information on the forest floor is very difficult right now without a lot of manual labor or a very expensive lidar system.”
“So it’s a groundbreaking, cost-effective device,” Baudreau said. “It won’t give you the same resolution data as lidar, but that’s just the first application. We hope that making VegSense open source for the environmental research community will push all the potential ways it can be developed.”
Northrop Grumman, Conservation International and Rice supported the research.
Video: https://youtu.be/dBHfxvhMChU
Story source:
Materials is provided Rice University. Originally written by Mike Williams. Note: Content can be edited for style and length.
