Help Identify Tree Stems from Drone Data

The Tree Mapping project uses crowd sourcing to accurate identify tree stems from laser scanning point cloud data acquired via UAV.  The project is a collaboration between the University of Heidelberg’s Institute of Geography and Institute of Geography and Geoecology (IFGG), Karlsruhe Institute of Technology (KIT) and uses data collected via SYSSIFOSS.

As the project explains, there is currently not an automated method of identifying single trees from the collected laser data.  The Tree Mapping project asks users to manually pinpoint single trees by scanning single pictures of profiles through the forest point clouds and marking the lower stem of trees.  Participation is simple and involves users marking (if any) the lower stems of tree profiles found in profiles of forest point clouds:

The profiles are color shaded based on the strength of the laser signal.  Some profiles contain no captured tree stems in which there is the option to note that no visible stem”

A profile with no stems visible.

A profile with no tree stems visible.

Each session serves up 20 profiles to process.  Each profiles takes between 1-5 seconds to complete, depending on the complexity and strength of the laser data captured.  Profiles with poor data quality and therefore no identifiable stems are quickly processed as no stems.  Profiles with strong laser data may have 5-10 easily visible tree stems to flag.

The laser data can be strong (left) or weak (right).

The laser data can be strong (left) or weak (right).

To get start visit 3D Micro-Mapping Tree Project and participate by signing in with a name of your choosing and clicking on the CAPTCHA.  You can come back to your profile and keep track of your participation by saving the URL information for your profile which is provided at the end of a session.

Related


Ad:


Like this article and want more?

Enter your email to receive the weekly GIS Lounge newsletter:

Advertising