4.02.05 - Remote sensing
Unit 4.02.05 is specifically concerned with research on remote sensing methods and applications for forest monitoring. Although remote sensing is used in many other units and working groups, this unit has the specific focus on bringing forward improved methods, tools, algorithms and techniques, which can support the research on forests on all levels. Therefore the scale of applications is very wide ranging from single tree asssessments to continental and global applications. Activities are carried out through informal discussions, meetings, and conferences (preferably jointly with other IUFRO Units), and by drawing attention to research needs and gaps in existing knowledge.
On the scale of large area monitoring with satellite data, current Earth observation (EO) missions such as the Sentinels acquire vast volume of data, e.g. for Sentinel-1 and 2 a new image every 5 days of almost every place on earth. By taking orbit overlaps into account, the time between consecutive images of the same region is reduced even further. Through high-quality georeferencing, it is possible to create consistent time series of satellite data making near real time applications and time series based classifications at a high spatial detail possible. Challenges can be found in the seasonality of the forests, irregular missing data or artefacts. Also, big data applications and the use of artificial intelligence for dealing with these vast amounts of data are currently in the focus of the research community.
On the scale of detailed monitoring, LiDAR systems play a significant role. They are employed from different platforms such as terrestrial scanners, handheld devices, drones and aerial surveys. Their fusion and combination with other data sets are current research topics of high relevance to local applications and also for the generation of reference data for multiple other uses. Using deep learing technology not only on image data, but on 3D point clouds is another important research issue.
- time series analysis
- meaningful fusion of different sensor type data
- near real time applications
- artificial intelligence for various applications
- deep learning for 3D point cloud analysis
- space-based LiDAR