Transfer learning in convolutional neural networks for fire scene classification in the Pantanal Biome (Brazil)
Authors:
- Jack Endrick Pastrana-MojicaJPJack Endrick Pastrana-Mojica
- Tássia Fraga-BelloliTFTássia Fraga-Belloli
- Pamela Boelter-HerrmannPBPamela Boelter-Herrmann
- Ana Júlia Magalhães-Gonçalves DiasADAna Júlia Magalhães-Gonçalves Dias
- Deyvis CanoDeyvis Cano
- Camila Souza-SilvaCSCamila Souza-Silva
- Daniel Capella-ZanottaDCDaniel Capella-Zanotta
Open Access
- 13 May 2026
- journal article
- Published by Universitat Politecnica de Valencia in Revista de Teledetección
- No. 68
Abstract
During 2024, wildfires in the Pantanal, Brazil, posed a significant threat to the world’s largest wetland. This study presents a scene classification approach using Sentinel-2 imagery from fire-affected and non-affected areas, based on deep learning techniques. First, a model was experimentally trained using the EuroSAT dataset to classify grasslands and herbaceous vegetation. Subsequently, a transfer learning technique was applied to images of the fire-affected Pantanal. Finally, the VGG-19 architecture was trained from scratch, without using pretrained parameters. Considering these three experiments, training with the EuroSAT dataset achieved a loss of 0.202 after 50 epochs. The model with transfer learning, using the weights learned during the experimental training, obtained a validation loss of 0.15 and an accuracy of 95%, whereas training from scratch reached a validation loss of 0.07 and an accuracy of 97%. The results indicate similar performance between the two strategies, demonstrating that in-domain training can be considered an effective method for training CNNs applied to satellite imagery, without compromising classification accuracy, even in the presence of significant differences in the contextual information of the images. This approach is presented as a methodological alternative for environmental monitoring in complex tropical regions, characterized by high spatial and spectral heterogeneity.Keywords
All Related Versions
This publication has 42 references indexed in Scilit:
- Analyzing wildfire patterns and climate interactions in Campania, Italy: A multi-sensor remote sensing studyEcological Informatics, 2025
- Changes, trends, and gaps in research dynamics after the megafires in the PantanalEnvironment, Development and Sustainability, 2024
- Comparison of diverse machine learning algorithms for forest fire susceptibility mapping in Antalya, TürkiyeAdvances In Space Research, 2024
- Using ensemble machine learning algorithm to predict forest fire occurrence probability in Madhya Pradesh and Chhattisgarh, IndiaAdvances In Space Research, 2023
- Assessment of fire resilience in subtropical wetlands using high spatial resolution imagesEnvironmental Monitoring and Assessment, 2022
- 2020 Pantanal’s widespread fire: short- and long-term implications for biodiversity and conservationBiodiversity and Conservation, 2021
- Characterizing Distribution of Forest Fires in Myanmar Using Earth Observations and Spatial Statistics ToolJournal of the Indian Society of Remote Sensing, 2019
- PatternNet: A benchmark dataset for performance evaluation of remote sensing image retrievalISPRS Journal of Photogrammetry and Remote Sensing, 2018
- Deep learning for visual understanding: A reviewNeurocomputing, 2016
- A simple semi-analytical model for remote estimation of chlorophyll-a in turbid waters: ValidationRemote Sensing of Environment, 2008