16 November 2023
This week, SWIFTT partner from the University of Bari Aldo Moro Bari Annalisa Appice presented the recently published study “SILVIA: An eXplainable Framework to Map Bark Beetle Infestation in Sentinel-2 Images” at the Recently Published Papers Track of the 3rd International Joint Conference on Learning & Reasoning (IJCLR) held in Bari, Italy.
The paper explores how machine learning methods can be used to map bark beetle infestation hotspots in Copernicus Sentinel 2 images by leveraging spectral vegetation indices and performing self-training. Another contribution of this study is the use of an eXplainable Artificial Intelligence (XAI) technique to strengthen the mapping results for exploring the effect of both spectral bands and spectral vegetation indices on decisions concerning forest patches with healthy and dying trees.
During her talk, Annalisa also presented SWIFTT’s main objectives and the results achieved by the research team from the University of Bari Aldo Moro.
Read the paper: G. Andresini, A. Appice and D. Malerba, "SILVIA: An eXplainable Framework to Map Bark Beetle Infestation in Sentinel-2 Images," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, doi: 10.1109/JSTARS.2023.3312521.
Learn more about the event on the link below