This research project focuses on the application of machine and deep learning methods for forestry applications. In particular, the main focus is forest growing stock prediction in the Friuli Venezia Giulia region (Italy), but the developed methods can be applied to produce estimations of biophysical forest attributes on any large territory.
This study will take into account different sources of data such as Forest inventory data from Nationla surveys, multispectral satellite images, climatic data and various environmental features collected through different services. Several methods will be applied to produce a forest-growing stock volume map, which will be useful to create management plans for forestry areas in the region.
Traditionally, the growing stock is considered an important indicator of forest health and productivity. The growing stock is estimated through forest inventory under which both qualitative and quantitative parameters are recorded to know the overall health of growing forests. So, we will produce the results that can be considered as a basis for evaluating current forest inventory, planning bio-economy, and sustainable forest management.
- Beatrice Portelli (AILAB-Udine)
- Mehdi Fasihi (AILAB-Udine)
- Luca Cadez (Uniud DI4A)
- Antonio Tomao (Uniud DI4A)
- Giorgio Alberti (Uniud DI4A)
- Giuseppe Serra (AILAB-Udine)