Automatic deforestation detection with deep learning: the case of Masoala National Park
Résumé
Large portions of territories are affected by deforestation. We aim to propose an approach based on convolutional neural networks to detect change in tropical forests. Our goal is to propose a model that requires minimal preprocessing and handcrafted features. We will test this approach with images from the Masoala National Park in Madagascar.
Domaines
Informatique [cs]| Origine | Fichiers produits par l'(les) auteur(s) |
|---|