Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data
Remotely sensed data are dominated by mixed land use and land cover (LULC) types. Spectral unmixing (SU) is a key technique that disentangles mixed pixels into constituent LULC types and their abundance fractions. While existing studies on deep learning (DL) for SU typically focus on single time-ste...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10416323/ |
_version_ | 1797302665210757120 |
---|---|
author | Jose Rodriguez-Ortega Rohaifa Khaldi Domingo Alcaraz-Segura Siham Tabik |
author_facet | Jose Rodriguez-Ortega Rohaifa Khaldi Domingo Alcaraz-Segura Siham Tabik |
author_sort | Jose Rodriguez-Ortega |
collection | DOAJ |
description | Remotely sensed data are dominated by mixed land use and land cover (LULC) types. Spectral unmixing (SU) is a key technique that disentangles mixed pixels into constituent LULC types and their abundance fractions. While existing studies on deep learning (DL) for SU typically focus on single time-step hyperspectral or multispectral data, our work pioneers SU using MODIS MS time series, addressing missing data with end-to-end DL models. Our approach enhances a long-short-term-memory-based model by incorporating geographic, topographic (geo-topographic), and climatic ancillary information. Notably, our method eliminates the need for explicit endmember extraction, instead learning the input–output relationship between mixed spectra and LULC abundances through supervised learning. Experimental results demonstrate that integrating spectral-temporal input data with geo-topographic and climatic information significantly improves the estimation of LULC abundances in mixed pixels. To facilitate this study, we curated a novel labeled dataset for Andalusia (Spain) with monthly MODIS MS time series at 460-m resolution for 2013. Named Andalusia MultiSpectral MultiTemporal Unmixing, this dataset provides pixel-level annotations of LULC abundances along with ancillary information. |
first_indexed | 2024-03-07T23:41:06Z |
format | Article |
id | doaj.art-28fa2fcc781c495cb33e09d179cfb08a |
institution | Directory Open Access Journal |
issn | 2151-1535 |
language | English |
last_indexed | 2024-03-07T23:41:06Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
spelling | doaj.art-28fa2fcc781c495cb33e09d179cfb08a2024-02-20T00:00:20ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01174626464510.1109/JSTARS.2024.335964710416323Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic DataJose Rodriguez-Ortega0https://orcid.org/0009-0000-6369-7229Rohaifa Khaldi1https://orcid.org/0000-0001-6224-2206Domingo Alcaraz-Segura2https://orcid.org/0000-0001-8988-4540Siham Tabik3https://orcid.org/0000-0003-4093-5356Department of Computer Science and Artificial Intelligence, Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Granada, Granada, SpainDepartment of Computer Science and Artificial Intelligence, Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Granada, Granada, SpainDepartment of Botany, Faculty of Science and with the iEcolab, Andalusian Institute for Earth System Research (IISTA-CEAMA), University of Granada, Granada, SpainDepartment of Computer Science and Artificial Intelligence, Andalusian Research Institute in Data Science and Computational Intelligence (DaSCI), University of Granada, Granada, SpainRemotely sensed data are dominated by mixed land use and land cover (LULC) types. Spectral unmixing (SU) is a key technique that disentangles mixed pixels into constituent LULC types and their abundance fractions. While existing studies on deep learning (DL) for SU typically focus on single time-step hyperspectral or multispectral data, our work pioneers SU using MODIS MS time series, addressing missing data with end-to-end DL models. Our approach enhances a long-short-term-memory-based model by incorporating geographic, topographic (geo-topographic), and climatic ancillary information. Notably, our method eliminates the need for explicit endmember extraction, instead learning the input–output relationship between mixed spectra and LULC abundances through supervised learning. Experimental results demonstrate that integrating spectral-temporal input data with geo-topographic and climatic information significantly improves the estimation of LULC abundances in mixed pixels. To facilitate this study, we curated a novel labeled dataset for Andalusia (Spain) with monthly MODIS MS time series at 460-m resolution for 2013. Named Andalusia MultiSpectral MultiTemporal Unmixing, this dataset provides pixel-level annotations of LULC abundances along with ancillary information.https://ieeexplore.ieee.org/document/10416323/Abundance estimationbidirectional long-short term memory (LSTM)climatic datadeep learning (DL)geo-topographic dataland use and land cover (LULC) |
spellingShingle | Jose Rodriguez-Ortega Rohaifa Khaldi Domingo Alcaraz-Segura Siham Tabik Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Abundance estimation bidirectional long-short term memory (LSTM) climatic data deep learning (DL) geo-topographic data land use and land cover (LULC) |
title | Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data |
title_full | Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data |
title_fullStr | Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data |
title_full_unstemmed | Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data |
title_short | Bidirectional Recurrent Imputation and Abundance Estimation of LULC Classes With MODIS Multispectral Time-Series and Geo-Topographic and Climatic Data |
title_sort | bidirectional recurrent imputation and abundance estimation of lulc classes with modis multispectral time series and geo topographic and climatic data |
topic | Abundance estimation bidirectional long-short term memory (LSTM) climatic data deep learning (DL) geo-topographic data land use and land cover (LULC) |
url | https://ieeexplore.ieee.org/document/10416323/ |
work_keys_str_mv | AT joserodriguezortega bidirectionalrecurrentimputationandabundanceestimationoflulcclasseswithmodismultispectraltimeseriesandgeotopographicandclimaticdata AT rohaifakhaldi bidirectionalrecurrentimputationandabundanceestimationoflulcclasseswithmodismultispectraltimeseriesandgeotopographicandclimaticdata AT domingoalcarazsegura bidirectionalrecurrentimputationandabundanceestimationoflulcclasseswithmodismultispectraltimeseriesandgeotopographicandclimaticdata AT sihamtabik bidirectionalrecurrentimputationandabundanceestimationoflulcclasseswithmodismultispectraltimeseriesandgeotopographicandclimaticdata |