Predicting Sagebrush Flowering with Machine Learning
Additional Funding Sources
This project was made possible by the NSF Idaho EPSCoR Program and by the National Science Foundation under Award No. OIA-1757324.
Abstract
The purpose of this project is to test if we can predict which sagebrush will flower by utilizing a combination of drone imagery, remote sensing, and machine learning. Data are collected via plotted drone flights at sites in which wildfires changed the landscape and affected sagebrush growth and development. These sites are analyzed to see how sagebrush recover from fire. The Drone flights, in combination with precisely geo-referenced locations of individual plants measured in the field, allow predictions of flowering using GIS related software.
Data collected in the field gives accurate positions, high resolution imagery, elevation maps, 3D images, and spectral imagery, which is useful for analyzing vegetation. These datasets are added to a GIS application. Tools within said application allow for filtering of sagebrush from other vegetation and the ground. The size and shape of the sagebrush, its UV reflectance, and data collection from the field of which shrubs are already flowering, will be the training data for a machine learning algorithm. The expected result from feeding this data is for the algorithm to be able to predict which sagebrush will flower and which will not to some degree.
Predicting Sagebrush Flowering with Machine Learning
The purpose of this project is to test if we can predict which sagebrush will flower by utilizing a combination of drone imagery, remote sensing, and machine learning. Data are collected via plotted drone flights at sites in which wildfires changed the landscape and affected sagebrush growth and development. These sites are analyzed to see how sagebrush recover from fire. The Drone flights, in combination with precisely geo-referenced locations of individual plants measured in the field, allow predictions of flowering using GIS related software.
Data collected in the field gives accurate positions, high resolution imagery, elevation maps, 3D images, and spectral imagery, which is useful for analyzing vegetation. These datasets are added to a GIS application. Tools within said application allow for filtering of sagebrush from other vegetation and the ground. The size and shape of the sagebrush, its UV reflectance, and data collection from the field of which shrubs are already flowering, will be the training data for a machine learning algorithm. The expected result from feeding this data is for the algorithm to be able to predict which sagebrush will flower and which will not to some degree.