Automated detection of burned areas in Costa Rica: a first approach
Αποθηκεύτηκε σε:
| Συγγραφείς: | , , , , |
|---|---|
| Μορφή: | artículo original |
| Κατάσταση: | Versión publicada |
| Ημερομηνία έκδοσης: | 2026 |
| Περιγραφή: | In Costa Rica despite diverse studies carried out by wildfires, collection data still is arduous fieldwork due to geographical conditions, there are zones where accessibility conditions prevent data collections. Satellite images are tools useful to study different zones to detect burned areas or their scars, but processing data by researchers requires too much time due to the number of files that need to be analyzed. We propose in this paper a framework based on machine learning and spectral index analysis to help burned area detection with efficient computational performance. Selecting as our study area in the Guanacaste Conservation Area, we obtained data from Sentinel-2 mission; we could detect the most probable zones affected by wildfire. Although this is a first step in the prevention of wildfire in protected zones, our results demonstrate the potential to develop a future robust detecting system. |
| Χώρα: | Portal de Revistas TEC |
| Ίδρυμα: | Instituto Tecnológico de Costa Rica |
| Repositorio: | Portal de Revistas TEC |
| Γλώσσα: | Español |
| OAI Identifier: | oai:ojs.pkp.sfu.ca:article/8504 |
| Διαθέσιμο Online: | https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/8504 |
| Λέξη-Κλειδί : | áreas quemadas detección clasificación aprendizaje automático aplicaciones prácticas de IA rendimiento computacional Burned areas detection classification machine learning practical applications of AI computational performance |