Open AccessDetection of forest roads in Sentinel-2 images using U-Net
Author(s) -
Øivind Due Trier,
Arnt-Børre Salberg,
Ragnvald Larsen,
Ole Torbjørn Nyvoll
Publication year - 2022
Publication title -
proceedings of the northern lights deep learning workshop
Language(s) - English
Resource type - Journals
ISSN - 2703-6928
DOI - 10.7557/18.6246
Subject(s) - false positive paradox , computer science , remote sensing , high resolution , cartography , environmental science , artificial intelligence , geography
This paper presents a new method for semi-automatic detection of nature interventions inSentinel-2 satellite images with 10 m spatial resolution. The Norwegian Environment Agency ismaintaining a map of undisturbed nature in Norway. U-Net was used for automated detection ofnew roads, as these are often the cause whenever the area of undisturbed nature is reduced. Themethod was able to detect many new roads, but with some false positives and possibly some falsenegatives (i.e., missing new roads). In conclusion, we have demonstrated that automated detection ofnew roads, for the purpose of updating the map of undisturbed nature, is possible. We have alsosuggested several improvements of the method to improve its usefulness.