CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification

The <italic>Cerrado</italic> faces increasing environmental pressures, necessitating accurate land use and land cover mapping despite challenges, such as class imbalance and visually similar categories. To address this, we present CerraData-4 MM, a multimodal dataset combining Sentinel-1...

Full description

Saved in:
Bibliographic Details
Main Authors: Mateus de Souza Miranda, Ronny Hansch, Valdivino Alexandre de Santiago Junior, Thales Sehn Korting, Erison Carlos dos Santos Monteiro
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11068119/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850078868561461248
author Mateus de Souza Miranda
Ronny Hansch
Valdivino Alexandre de Santiago Junior
Thales Sehn Korting
Erison Carlos dos Santos Monteiro
author_facet Mateus de Souza Miranda
Ronny Hansch
Valdivino Alexandre de Santiago Junior
Thales Sehn Korting
Erison Carlos dos Santos Monteiro
author_sort Mateus de Souza Miranda
collection DOAJ
description The <italic>Cerrado</italic> faces increasing environmental pressures, necessitating accurate land use and land cover mapping despite challenges, such as class imbalance and visually similar categories. To address this, we present CerraData-4 MM, a multimodal dataset combining Sentinel-1 synthetic aperture radar and Sentinel-2 multispectral imagery with 10 m spatial resolution. The dataset includes two hierarchical classification levels with seven and 14 classes, respectively, focusing on the diverse <italic>Bico do Papagaio</italic> ecoregion. We benchmark two models trained on CerraData-4 MM, employing a visual transformer-based architecture and a convolutional-based architecture. The ViT achieves superior performance in multimodal scenarios, with the highest macro F1-score of 57.60&#x0025; and a mean Intersection over Union of 49.05&#x0025; at the first hierarchical level. Both models struggle with minority classes, particularly at the second hierarchical level, where U-Net&#x2019;s performance drops to an F1-score of 18.16&#x0025;. Weighted loss improves representation for underrepresented classes but reduces overall accuracy, underscoring the trade-off in weighted training. CerraData-4 MM offers a challenging benchmark for advancing deep learning models to handle class imbalance and multimodal data fusion.
format Article
id doaj-art-2b8feb9eac7f4d7aa0354d4ef727a5bb
institution DOAJ
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-2b8feb9eac7f4d7aa0354d4ef727a5bb2025-08-20T02:45:26ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118180311804110.1109/JSTARS.2025.358580511068119CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover ClassificationMateus de Souza Miranda0https://orcid.org/0000-0001-7887-4048Ronny Hansch1https://orcid.org/0000-0002-2936-6765Valdivino Alexandre de Santiago Junior2https://orcid.org/0000-0002-4277-021XThales Sehn Korting3https://orcid.org/0000-0002-0876-0501Erison Carlos dos Santos Monteiro4https://orcid.org/0000-0003-0193-9450Laborat&#x00F3;rio de Intelig&#x00EA;ncia ARtificial para Aplica&#x00E7;&#x00F5;es AeroEspaciais e Ambientais (LIAREA), Programa de P&#x00F3;s-Gradua&#x00E7;&#x00E3;o em Computa&#x00E7;&#x00E3;o Aplicada (PGCAP), Instituto Nacional de Pesquisas Espaciais (INPE), S&#x00E3;o Jos&#x00E9; dos Campos, BrazilGerman Aerospace Center (DLR), Oberpfaffenhofen, GermanyLaborat&#x00F3;rio de Intelig&#x00EA;ncia ARtificial para Aplica&#x00E7;&#x00F5;es AeroEspaciais e Ambientais (LIAREA), Programa de P&#x00F3;s-Gradua&#x00E7;&#x00E3;o em Computa&#x00E7;&#x00E3;o Aplicada (PGCAP), Instituto Nacional de Pesquisas Espaciais (INPE), S&#x00E3;o Jos&#x00E9; dos Campos, BrazilLaborat&#x00F3;rio de Intelig&#x00EA;ncia ARtificial para Aplica&#x00E7;&#x00F5;es AeroEspaciais e Ambientais (LIAREA), Programa de P&#x00F3;s-Gradua&#x00E7;&#x00E3;o em Computa&#x00E7;&#x00E3;o Aplicada (PGCAP), Instituto Nacional de Pesquisas Espaciais (INPE), S&#x00E3;o Jos&#x00E9; dos Campos, BrazilLaborat&#x00F3;rio de Intelig&#x00EA;ncia ARtificial para Aplica&#x00E7;&#x00F5;es AeroEspaciais e Ambientais (LIAREA), Programa de P&#x00F3;s-Gradua&#x00E7;&#x00E3;o em Computa&#x00E7;&#x00E3;o Aplicada (PGCAP), Instituto Nacional de Pesquisas Espaciais (INPE), S&#x00E3;o Jos&#x00E9; dos Campos, BrazilThe <italic>Cerrado</italic> faces increasing environmental pressures, necessitating accurate land use and land cover mapping despite challenges, such as class imbalance and visually similar categories. To address this, we present CerraData-4 MM, a multimodal dataset combining Sentinel-1 synthetic aperture radar and Sentinel-2 multispectral imagery with 10 m spatial resolution. The dataset includes two hierarchical classification levels with seven and 14 classes, respectively, focusing on the diverse <italic>Bico do Papagaio</italic> ecoregion. We benchmark two models trained on CerraData-4 MM, employing a visual transformer-based architecture and a convolutional-based architecture. The ViT achieves superior performance in multimodal scenarios, with the highest macro F1-score of 57.60&#x0025; and a mean Intersection over Union of 49.05&#x0025; at the first hierarchical level. Both models struggle with minority classes, particularly at the second hierarchical level, where U-Net&#x2019;s performance drops to an F1-score of 18.16&#x0025;. Weighted loss improves representation for underrepresented classes but reduces overall accuracy, underscoring the trade-off in weighted training. CerraData-4 MM offers a challenging benchmark for advancing deep learning models to handle class imbalance and multimodal data fusion.https://ieeexplore.ieee.org/document/11068119/Cerradodeep learning (DL)hierarchical level of classesland use and land cover (LULC) classificationsemantic segmentation
spellingShingle Mateus de Souza Miranda
Ronny Hansch
Valdivino Alexandre de Santiago Junior
Thales Sehn Korting
Erison Carlos dos Santos Monteiro
CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Cerrado
deep learning (DL)
hierarchical level of classes
land use and land cover (LULC) classification
semantic segmentation
title CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification
title_full CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification
title_fullStr CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification
title_full_unstemmed CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification
title_short CerraData-4 MM: A Multimodal Benchmark Dataset on Cerrado for Land Use and Land Cover Classification
title_sort cerradata 4 mm a multimodal benchmark dataset on cerrado for land use and land cover classification
topic Cerrado
deep learning (DL)
hierarchical level of classes
land use and land cover (LULC) classification
semantic segmentation
url https://ieeexplore.ieee.org/document/11068119/
work_keys_str_mv AT mateusdesouzamiranda cerradata4mmamultimodalbenchmarkdatasetoncerradoforlanduseandlandcoverclassification
AT ronnyhansch cerradata4mmamultimodalbenchmarkdatasetoncerradoforlanduseandlandcoverclassification
AT valdivinoalexandredesantiagojunior cerradata4mmamultimodalbenchmarkdatasetoncerradoforlanduseandlandcoverclassification
AT thalessehnkorting cerradata4mmamultimodalbenchmarkdatasetoncerradoforlanduseandlandcoverclassification
AT erisoncarlosdossantosmonteiro cerradata4mmamultimodalbenchmarkdatasetoncerradoforlanduseandlandcoverclassification