Please note that this session was withdrawn and is no longer available in the respective programme. This withdrawal might have been the result of a merge with another session.

NH3.3

EDI
Artificial intelligence for Event-Based Landslide Detection

This session will cover both new approaches and state-of-the-art Artificial intelligence techniques on Remote Sensing (RS) data for creating event-based landslides inventories and for updating existing ones.
Landslides are geomorphological phenomena with a high potential of causing heavy economical and human losses. Natural calamities, such as earthquakes, typhoons, and extreme rainfall events frequently trigger multiple landslide occurrences, other factors such as human interventions also also causing several landslides in urban and rural areas around the world. Furthermore, recently, the effects of climate change are increasing the temporal and spatial probability of landslide events.
Several case studies have demonstrated how well-structured hazard and/or risk assessment models are fundamental for long/short-term risk reduction.
Event-based landslide inventories are critical for the development and validation of reliable susceptibility hazard and risk models, as well as for the understanding of the event itself. These inventories are also the basis of validating the results of machine /deep learning models outputs.
In this session, submissions are encouraged related to all landslide types, from rock falls to rapid debris flows, slow-moving deep-seated landslides. Submissions related to All Artificial Intelligence methods and algorithms are welcome. We encourage the use of a regional scale analysis for landslide detection and applications for the establishment of multi-temporal inventories.
Contributions can be related to investigating data processing, fusion, and data manipulation as well as model tuning practices. Contributions are welcomed which aims at the evaluation of the quality of landslide detection; the comparison of the performance of different segmentation models; and investigations of the potential for the exploitation of new or emerging technologies e.g., computational, Earth observation technologies, in order to improve our ability to create event-based landslide inventories.
We believe that your contributions will greatly boost the quality and the advancements in this field, bridging the existing research gaps.

Co-organized by