Centre de documentation
scientifique

Large-scale burn severity mapping in multispectral imagery using deep semantic segmentation models / Xikun Hu in ISPRS Journal of photogrammetry and remote sensing, vol 196 (February 2023)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 196 (February 2023) . - pp 228 - 240 Titre : Large-scale burn severity mapping in multispectral imagery using deep semantic segmentation models Type de document : Article/Communication Auteurs : Xikun Hu, Auteur ; Puzhao Zhang, Auteur ; Yifang Ban, Auteur Année de publication : 2023 Article en page(s) : pp 228 - 240 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] carte thématique [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] dommage [Termes IGN] image Landsat-ETM+ [Termes IGN] image Landsat-OLI [Termes IGN] image Landsat-TM [Termes IGN] image multibande [Termes IGN] image Sentinel-MSI [Termes IGN] incendie de forêt [Termes IGN] jeu de données localisées [Termes IGN] segmentation sémantique [Termes IGN] surveillance forestière [Termes IGN] zone sinistrée Résumé : (auteur) Nowadays Earth observation satellites provide forest fire authorities and resource managers with spatial and comprehensive information for fire stabilization and recovery. Burn severity mapping is typically performed by classifying bi-temporal indices (e.g., dNBR, and RdNBR) using thresholds derived from parametric models incorporating field-based measurements. Analysts are currently expending considerable manual effort using prior knowledge and visual inspection to determine burn severity thresholds. In this study, we aim to employ highly automated approaches to provide spatially explicit damage level estimates. We first reorganize a large-scale Landsat-based bi-temporal burn severity assessment dataset (Landsat-BSA) by visual data cleaning based on annotated MTBS data (approximately 1000 major fire events in the United States). Then we apply state-of-the-art deep learning (DL) based methods to map burn severity based on the Landsat-BSA dataset. Experimental results emphasize that multi-class semantic segmentation algorithms can approximate the threshold-based techniques used extensively for burn severity classification. UNet-like models outperform other region-based CNN and Transformer-based models and achieve accurate pixel-wise classification results. Combined with the online hard example mining algorithm to reduce class imbalance issue, Attention UNet achieves the highest mIoU (0.78) and the highest Kappa coefficient close to 0.90. The bi-temporal inputs with ancillary spectral indices work much better than the uni-temporal multispectral inputs. The restructured dataset will be publicly available and create opportunities for further advances in remote sensing and wildfire communities. Numéro de notice : A2023-122 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2022.12.026 Date de publication en ligne : 11/01/2023 En ligne : https://doi.org/10.1016/j.isprsjprs.2022.12.026 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102498 [article]