UH Hilo researchers develop AI software to find the invasive species miconia in Maui forests

Equipment on outside of helicopter records visual surveys of Maui forests; UH Hilo researchers have trained artificial intelligence software to recognize invasive miconia in the thicket.

Yellow helicopter on landing pad with pilot and passenger. Cameras are mounted under helicopter.
Roberto Rodriguez from the Spatial Data Analysis and Visualization laboratory at UH Hilo preps for take off on a test flight to collect aerial imagery of East Maui forests. Rodriguez has developed an artificial intelligence computer program that identifies invasive miconia from photographs, saving manual field crews hours of work. (Photo: Adam Knox/Maui Invasive Species Committee)

By Susan Enright.

Ryan Perroy pictured
Ryan Perroy

Ryan Perroy and Roberto Rodriguez with the Spatial Data Analysis and Visualization Laboratory at the University of Hawaiʻi at Hilo have developed a process to collect and analyze aerial footage of Maui’s forests to identify the invasive species miconia.

Perroy is a UH Hilo professor of geography and environmental science at the College of Arts and Sciences, and also serves as director of the SDAV Laboratory. Rodriguez is an assistant professor of aeronautical sciences based at UH Hilo’s College of Agriculture, Forestry and Natural Resource Management.

After years of hanging out of helicopters to search low-elevation forests for the telltale green and purple on miconea leaves, an activity that often left investigators with sore necks and sometimes nausea, the Maui Invasive Species Committee has turned to Perroy and his lab that uses computer software to identify invasive plants from aerial images. Perroy’s research group has had great success with surveying native forests to find trees diseased with Rapid ʻŌhiʻa Death.

The process still involves collecting aerial images via helicopter, but instead of relying on human spotters, Perroy and his team developed a way to mount digital cameras on the outside of the helicopter that capture imagery to be analyzed back at the SDAV Lab.

The cameras and equipment record the location, orientation, and speed of the helicopter. Perroy’s co-researcher Rodriguez has trained the artificial intelligence software to recognize miconia.

“The software can recognize a single leaf poking through the forest,” Perroy explains.

Arial image of forest with black circular outlines of locations of miconia.
The artificial intelligence computer program identifies miconia in aerial photographs using visual cues such as leaf size and shape. Above, miconia plants are circled based on computer identification. Each plant is linked to global positioning system or GPS coordinates giving crews a location so they can find and remove the plant. (Photo: Spatial Data Analysis and Visualization Lab/UH Hilo)
Several photos of invasive miconia.
Researchers at UH Hiloʻs Spatial Data Analysis and Visualization Laboratory are surveying non-native weeds, including miconia (Miconia calvescens) on Maui. Miconia is an invasive noxious weed found throughout the main Hawaiian Islands that shades out native plant species and quickly takes over a forest to create an invasive monotypic forest. (Photos: Spatial Data Analysis and Visualization Lab/UH Hilo)

This use of artificial intelligence, or AI, excels at repetitive and redundant tasks, makes it faster and easier for investigations to move forward, increases workflow efficiency, and speeds up data analysis.

Back at the lab, the computer then assigns global positioning systems or GPS coordinates to the plant using data collected during the flight. The team then creates visual maps — a form of data visualization — that are super easy for anyone to read.

The information will help guide future work to ensure the long-term health of Maui’s forested watersheds.

Read full story at The Maui News.


Story by Susan Enright, a public information specialist for the Office of the Chancellor and editor of UH Hilo Stories. She received her bachelor of arts in English and certificate in women’s studies from UH Hilo.

Share this story