This page contains the list of accepted presentations.You can use the controls in the side panel to order the list in various ways, and to toggle all the abstracts on or off, at once.

Downloadable material is attached to some of the presentations. The material is made available with the express consent of the respective presenters. All rights to the material are reserved by the the respective authors. If you want to, e.g., share, modify or build upon the work, please consult the respective authors for consent. Presentations with attached material are marked with the Downloadable tag.

If you are a presenter, you can view the presentation guidelines.

Disney, Mathias New approaches to TLS registration and information extraction from path analysis Poster
Mathias Disney1,2, Matheus Boni Vicari1, Phil Wilkes1,2 and Wanxin Yang1
(1) Dept. of Geography, University College London(2) NERC National Centre for Earth Observation

Terrestrial laser scanning data have been widely collected from forests and individual trees. These data provide 3D detail that is being used to address a range of ecological questions. As more TLS data are collected with different instruments and protocols and from a range of forest environments, particular challenges have emerged in extracting the most detailed 3D structural information from the resulting point clouds. One important (and often very time-consuming) challenge is accurate co-registration of many individual TLS scans into a single point clouds. This is typically done using target-based, SLAM and now IMU-based methods. Automated co-registration is possible, but remains a challenge for many instruments and natural environments. A second challenge is extracting architecture of individual trees from the resulting co-registered TLS point clouds. Quantitative structural model (QSM) approaches have been applied widely and successfully for dense, high accuracy point clouds. However, the influence of leaves, occlusion and other acquisition properties limit the resulting accuracy. We present preliminary results of a topology extraction approach that might potentially be used to address both co-registration and tree architecture simultaneously. The method was developed for separation of leaf and wood material, but the resulting iterative tree skeleton extraction potentially allows co-registration and volume enclosure to be considered as part of the same process. Combining this kind of approach in an AI/ML framework ought to allow the process to perform better over time by enabling learning and adapting, so that each application helps the next to be better.