Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Forest Biomass Mapping with Terrestrial Hyperspectral Imaging for Wildfire Risk Monitoring

Nathaniel Hanson, Sarvesh Prajapati, James Tukpah, Yash Mewada, Taşkın Padır

Abstract

With the rapid increase in wildfires in the past decade, it has become necessary to detect and predict these disasters to mitigate losses to ecosystems and human lives. In this paper, we present a novel solution -- Hyper-Drive3D -- consisting of snapshot hyperspectral imaging and LiDAR, mounted on an Unmanned Ground Vehicle (UGV) that identifies areas inside forests at risk of becoming fuel for a forest fire. This system enables more accurate classification by analyzing the spectral signatures of forest vegetation. We conducted field trials in a controlled environment simulating forest conditions, yielding valuable insights into the system's effectiveness. Extensive data collection was also performed in a dense forest across varying environmental conditions and topographies to enhance the system's predictive capabilities for fire hazards and support a risk-informed, proactive forest management strategy. Additionally, we propose a framework for extracting moisture data from hyperspectral imagery and projecting it into 3D space.

Forest Biomass Mapping with Terrestrial Hyperspectral Imaging for Wildfire Risk Monitoring

Abstract

With the rapid increase in wildfires in the past decade, it has become necessary to detect and predict these disasters to mitigate losses to ecosystems and human lives. In this paper, we present a novel solution -- Hyper-Drive3D -- consisting of snapshot hyperspectral imaging and LiDAR, mounted on an Unmanned Ground Vehicle (UGV) that identifies areas inside forests at risk of becoming fuel for a forest fire. This system enables more accurate classification by analyzing the spectral signatures of forest vegetation. We conducted field trials in a controlled environment simulating forest conditions, yielding valuable insights into the system's effectiveness. Extensive data collection was also performed in a dense forest across varying environmental conditions and topographies to enhance the system's predictive capabilities for fire hazards and support a risk-informed, proactive forest management strategy. Additionally, we propose a framework for extracting moisture data from hyperspectral imagery and projecting it into 3D space.

Paper Structure

This paper contains 15 sections, 7 equations, 6 figures.

Table of Contents

  1. INTRODUCTION
  2. RELATED WORK
  3. Practitioner Approaches
  4. Wildfire Quantification
  5. Autonomous Systems
  6. TECHNICAL APPROACH
  7. Imaging System
  8. Estimating Fuel Moisture Content
  9. EXPERIMENTATION AND DATASET
  10. Preliminary Data Collection
  11. Final Data Collection
  12. Moisture Content
  13. Results
  14. Conclusion
  15. Future Work

Figures (6)

  • Figure 2: Hyper-Drive3D system platform. The system contains a multi-modal platform for visible-shortwave infrared hyperspectral imaging of natural environments with continuous reflectance calibration provided by solar irradiance measurements. The system is mounted to a Warthog Unmanned Ground Vehicle. (Inset) Quantum efficiency and band spacing of the snapshot imaging array.
  • Figure 3: Data visualization from field collection. LiDAR points projected on the RGB image (left), registered hyperspectral cube ($816 \times 684 \times 36$) (center) and spectral profile of a selected pixel from the RGB Image (Top Right: profile from VNIR; Top Left: from SWIR).
  • Figure 4: The map illustrates a dense point-cloud map created by LeGO-LOAM 8594299 and shows the trajectory followed by the Warthog Unmanned Ground Vehicle (UGV) during data collection. The point density of the rendered maps shows a sufficient ability to differentiate individual trees and plants.
  • Figure 5: Visualization of all five routes for data collection at Olin (Image courtesy of Google Earth).
  • Figure 6: System diagram for generating registered point cloud from Raw Hyperspectral Datacube.
  • ...and 1 more figures