Drone-Image-Based Method of Estimating Forest-Fire Fuel Loads

Abstract

A method of estimating forest-fire fuel loads was developed using drones to collect information about the height and diameter-at-breast-height (DBH) of individual trees. It was conducted for forest fire prevention monitoring (Control, 20% thinned, and 40% thinned area) located in Goseong-gun, Gangwon-do. Object-based images and 3D-model red/green/blue band characteristics were superimposed to select and extract individual trees. A digital crown height model was developed based on the difference between the heights of digital surface and terrain models. In addition, the DBH was estimated based on the crown area. The 40%-thinned area exhibited the highest accuracy (95%) for extracting individual trees, and the difference between the field-survey and drone-image heights was in the range of 0.64-2.02 m. The goodness-of-fit of the DBH-crown area model was 0.61. The difference between the imageand field-survey-based forest-fire fuel loads ranged from -1.20 to 0.40 ton/ha.

Automated Segmentation of Individual Tree Structures Using Deep Learning over LiDAR Point Cloud Data

“…The structure of the trees needs to be segmented first. The trunk is used to estimate the wood volume, and the branch is utilized to calculate the amount of the fuel and biomass in forest fires [5,11,12]. However, the degree of difficulty in dividing tree structures can vary depending on the tree species and stand structure.…”