NINFA
Deep learning-based methods for the semantic segmentation of tree-like vegetation: Arecaceae, Pinus, Platanus & Celtis Australis
The aim of this project is to use Deep Learning techniques to identify and predict Green Infrastructure Ecosystem Services from high resolution satellite images, focusing specifically on the Mediterranean basin and on a strategic selection of plant taxa: the family Arecaceae, the genera Pinus and Platanus, and the species Celtis australis.
The core of the project involves the identification and classification of these plant taxa using advanced image segmentation algorithms. Deep Learning techniques will be instrumental in creating mathematical models for the assessment and prediction of four Ecosystem Services. These services, aligned with the Common International Classification of Ecosystem Services (CICES), include atmospheric regulation, thermal and moisture regulation, erosion data control and the enhancement of physical and experiential interaction with the natural environment.
Funding Entity
