Publicación: Self-Supervised Learning on Small In-Domain Datasets Can Overcome Supervised Learning in Remote Sensing
Cargando...
Fecha
2024
Editor/a
Director/a
Tutor/a
Coordinador/a
Prologuista
Revisor/a
Ilustrador/a
Derechos de acceso
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editor
IEEE
Resumen
The availability of high-resolution satellite images has accelerated the creation of new datasets designed to tackle broader remote sensing (RS) problems. Although popular tasks, such as scene classification, have received significant attention, the recent release of the Land-1.0 RS dataset marks the initiation of endeavors to estimate land-use and land-cover (LULC) fraction values per RGB satellite image. This challenging problem involves estimating LULC composition, i.e., the proportion of different LULC classes from satellite imagery, with major applications in environmental monitoring, agricultural/urban planning, and climate change studies. Currently, supervised deep learning models—the state-of-the-art in image classification—require large volumes of labeled training data to provide good generalization. To face the challenges posed by the scarcity of labeled RS data, self-supervised learning (SSL) models have recently emerged, learning directly from unlabeled data by leveraging the underlying structure. This is the first article to investigate the performance of SSL in LULC fraction estimation on RGB satellite patches using in-domain knowledge. We also performed a complementary analysis on LULC scene classification. Specifically, we pretrained Barlow Twins, MoCov2, SimCLR, and SimSiam SSL models with ResNet-18 using the Sentinel2GlobalLULC small RS dataset and then performed transfer learning to downstream tasks on Land-1.0. Our experiments demonstrate that SSL achieves competitive or slightly better results when trained on a smaller high-quality in-domain dataset of 194 877 samples compared to the supervised model trained on ImageNet-1k with 1 281 167 samples. This outcome highlights the effectiveness of SSL using in-distribution datasets, demonstrating efficient learning with fewer but more relevant data.
Descripción
The registered version of this article, first published in “IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17", is available online at the publisher's website: IEEE, https://doi.org/10.1109/JSTARS.2024.3421622
La versión registrada de este artículo, publicado por primera vez en “IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17", está disponible en línea en el sitio web del editor: IEEE, https://doi.org/10.1109/JSTARS.2024.3421622
Categorías UNESCO
Palabras clave
Task analysis, Estimation, Scene classification, Satellites, Satellite Images, Deep learning, Computational modeling, Climate change, Remote sensing, Self-supervised learning, Land use planning
Citación
A. J. Sanchez-Fernandez, S. Moreno-Álvarez, J. A. Rico-Gallego and S. Tabik, "Self-Supervised Learning on Small In-Domain Datasets Can Overcome Supervised Learning in Remote Sensing," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17, pp. 12797-12810, 2024, doi: 10.1109/JSTARS.2024.3421622
Centro
Facultades y escuelas::E.T.S. de Ingeniería Informática
Departamento
Lenguajes y Sistemas Informáticos