By integrating aerial and ground-based mobile mapping sensors and systems, a team of Purdue digital forestry researchers uses state-of-the-art technology to locate, count and measure over a thousand trees in a matter of hours.
“Machines count and measure every tree – it’s not modeling, it’s a real forest inventory,” said Songlin Faye, chairman of the dean’s remote monitoring and professor of forestry and natural resources and leader of the Digital Forest Initiative. farm of Purdue University. “This is a revolutionary development on our path to using technology to quickly and accurately describe the global forest ecosystem, which would improve our ability to prevent forest fires, detect diseases, perform accurate carbon counts and make informed management decisions. forests.
The technology uses manned aircraft, drones and backpack-mounted systems. The systems integrate cameras with light and range detection devices, or LiDAR, together with navigation sensors, including integrated global navigation satellite systems (GNSS) and inertial navigation systems (INS). A Purdue team led by Ayman Habib, Professor Thomas A. Page of Civil Engineering and head of the Purdue Digital Photogrammetry Research Group, which co-led the project with Fei, designed and built the systems.
“Different parts of the systems take advantage of the synergistic characteristics of the acquired data to determine which component has the most accurate information for a given data point,” Habib said. “This is how we can integrate small and large-scale information. One platform could not do that. We had to find a way for multiple platforms and sensors – providing different types of information – to work together. This gives a complete picture with extremely high resolution. Fine details are not lost. “
A machine learning algorithm developed by the data analysis team is as important as the personalized autonomous vehicles they create. The results of a study using their technology are described in detail in an article published in the journal Remote Sensing.
“This system collects a variety of information for each tree, including height, trunk diameter and branch information,” Habib said. “In addition to this information, we maintain accurate location and time markers for acquired features.”
The result is like giving a person the glasses he or she needs. What was once obscure and uncertain is becoming clear. Their vision improves, and so does their understanding of what they see.
LiDAR works like radar, but uses laser light as a signal. LiDAR sensors estimate the range between the scanning system and objects, using the time required for the signal to travel to objects and back to the sensor. On drones, airplanes or satellites, measurements are taken from above the canopy of trees, and on moving vehicles or backpacks, measurements are made from below the canopy. Air systems have continuous access to GNSS signals to accurately determine the location and orientation of the sensor after GNSS / INS integration and provide reasonable resolution. Ground-based systems, on the other hand, provide more detail and finer resolution, while suffering from potential GNSS signal interruptions, Habib said.
“This multi-platform system and processing framework takes the best of each to provide both fine detail and high positioning accuracy,” he said.
For example, if the backpack is in an area with poor access to GNSS signals, a drone can intervene and put that data in the right place, he said.
“This is a breakthrough in the application of new geomatic tools to forestry,” Faye said. “It solves a real and urgent challenge in areas like agriculture and transport, but it is also an amazing engineering and science that would be applied outside of one arena.”
Because the different platforms work together, the system also identifies data points from each that equate to the same tree feature. Ultimately, this could correlate with what the data above the canopy means in terms of what is happening under the canopy, Habib said. This would be a huge jump in the speed and area of forest that can be covered.