Student information

MSc thesis topic: Matching tree pointclouds from different LiDAR systems based on their tree structure

Light detection and ranging (LiDAR) has been proposed as a suitable technique for mapping forest. A number of parameters can be derived from these 3D maps such as tree height, diameter at breast height (DBH), and even tree volume providing an objective, practical and cost-effective solution for forest inventories.

LiDAR can be operated in airborne configuration (Airborne laser scanning (ALS) or in a terrestrial setup (TLS). TLS measures forests from below canopy and offers a much more detailed description of the individual trees and stands. In addition to this, traditional forest inventory also provide complement information about the forest, however, some measurements are subject to the instrument’s accuracy and the expert’s experience. These factors create uncertainty over the measured value. TLS can provide another, more detailed, value, which can improve the inventory.
For this research project, a we provide the basic script for matching tree pointclouds (in R), and two datasets: an UAV and TLS tree pointcloud datasets, co-registered and with segmented trees.

Objectives and Research questions

  • Understand the importance of a good quality alignment of trees for tree inventory based on different LiDAR systems.
  • Compare individual tree pointclouds from different LiDAR systems based on their tree characteristics.
  • Improve an existing algorithm to match tree pointclouds from different LiDAR systems based on the location and provide with (un)certainties based on their tree parameters.

Requirements

  • Basic understanding of LIDAR technology and their use in forests
  • Programming skills (R)

Literature and information

Expected reading list before starting the thesis research

  • Choose at least 5 research papers from the literature list above.

Theme (s): Modelling & visualisation