Titre:	Performance analysis of deep convolutional autoencoders with different patch sizes for change detection from burnt areas
Auteur(s):	Bem, Pablo Pozzobon de Carvalho Júnior, Osmar Abílio de Carvalho, Osmar Luiz Ferreira de Gomes, Roberto Arnaldo Trancoso Guimarães, Renato Fontes
metadata.dc.identifier.orcid:	https://orcid.org/ 0000-0003-3868-8704 https://orcid.org/ 0000-0002-0346-1684 https://orcid.org/ 0000-0002-5619-8525 https://orcid.org/ 0000-0003-4724-4064 https://orcid.org/ 0000-0002-9555-043X
Assunto::	Aprendizagem profunda Redes neurais (Computação) Classificação Fogo Imagem multitemporal
Date de publication:	11-aoû-2020
Editeur:	MDPI
Référence bibliographique:	BEM, Pablo Pozzobon de et al. Performance analysis of deep convolutional autoencoders with different patch sizes for change detection from burnt areas. Remote Sensing, v. 12, n. 16, 2576, 2020. DOI: https://doi.org/10.3390/rs12162576. Disponível em: https://www.mdpi.com/2072-4292/12/16/2576. Acesso em: 16 maio 2022.
Abstract:	Fire is one of the primary sources of damages to natural environments globally. Estimates show that approximately 4 million km2 of land burns yearly. Studies have shown that such estimates often underestimate the real extent of burnt land, which highlights the need to find better, state-of-the-art methods to detect and classify these areas. This study aimed to analyze the use of deep convolutional Autoencoders in the classification of burnt areas, considering di erent sample patch sizes. A simple Autoencoder and the U-Net and ResUnet architectures were evaluated. We collected Landsat 8 OLI+ data from three scenes in four consecutive dates to detect the changes specifically in the form of burnt land. The data were sampled according to four di erent sampling strategies to evaluate possible performance changes related to sampling window sizes. The training stage used two scenes, while the validation stage used the remaining scene. The ground truth change mask was created using the Normalized Burn Ratio (NBR) spectral index through a thresholding approach. The classifications were evaluated according to the F1 index, Kappa index, and mean Intersection over Union (mIoU) value. Results have shown that the U-Net and ResUnet architectures offered the best classifications with average F1, Kappa, and mIoU values of approximately 0.96, representing excellent classification results. We have also verified that a sampling window size of 256 by 256 pixels offered the best results.
Licença::	© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Collection(s) :	Artigos publicados em periódicos e afins

Tous les documents dans DSpace sont protégés par copyright, avec tous droits réservés.