Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics

Remote sensing has revolutionized crop mapping and monitoring, providing valuable insights for sustainable agricultural practices. However, successfully implementing remote sensing-based crop type identification in tropical regions remains challenging. Unlike temperate regions, tropical areas benefi...

Full description

Saved in:
Bibliographic Details
Main Authors: Laura Elena Cué La Rosa, Dario Augusto Borges Oliveira, Raul Queiroz Feitosa
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:International Journal of Applied Earth Observations and Geoinformation
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1569843225002638
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849240761120325632
author Laura Elena Cué La Rosa
Dario Augusto Borges Oliveira
Raul Queiroz Feitosa
author_facet Laura Elena Cué La Rosa
Dario Augusto Borges Oliveira
Raul Queiroz Feitosa
author_sort Laura Elena Cué La Rosa
collection DOAJ
description Remote sensing has revolutionized crop mapping and monitoring, providing valuable insights for sustainable agricultural practices. However, successfully implementing remote sensing-based crop type identification in tropical regions remains challenging. Unlike temperate regions, tropical areas benefit from favorable weather conditions that support diverse land management practices, resulting in complex crop dynamics that are difficult to model. Additionally, frequent cloud cover in tropical regions limits the use of optical data during extended periods of the year, making SAR (Synthetic Aperture Radar) a more attractive alternative. Traditional models like Conditional Random Fields (CRFs) have been useful in classifying crop types using spatial and temporal contexts. However, these models often use fixed inputs, optimizing them towards a specific task, and fail to consider the CRFs’ feedback for end-to-end learning features and the spatiotemporal dependencies. This work addresses this gap by introducing an end-to-end hybrid model that combines deep learning with CRFs to incorporate prior knowledge-based modeling of crop dynamics. The proposed framework integrates a backbone encoder–decoder to capture spatial and temporal contexts, with a CRF block that models temporal dynamics by accounting for label dependencies between adjacent epochs. This block facilitates the integration of both data-driven and expert-domain temporal constraints, allowing the framework to adapt to local agricultural practices and enhance crop dynamics modeling. We evaluate the framework using multi-temporal SAR image sequences from two municipalities in Brazil, comparing the effectiveness of learned versus prior-knowledge temporal constraints. The results show improvements of up to 30% in per-class F1 score and 12% in average F1 score compared to a baseline model that excludes temporal dependencies. These findings highlight the value of incorporating prior knowledge-driven temporal constraints into crop mapping models.
format Article
id doaj-art-dc2b2cb074744021af0ae81c8711e3d2
institution Kabale University
issn 1569-8432
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series International Journal of Applied Earth Observations and Geoinformation
spelling doaj-art-dc2b2cb074744021af0ae81c8711e3d22025-08-20T04:00:27ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322025-08-0114210461610.1016/j.jag.2025.104616Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamicsLaura Elena Cué La Rosa0Dario Augusto Borges Oliveira1Raul Queiroz Feitosa2Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, 6708 PB Wageningen, The Netherlands; Corresponding author.School of Applied Mathematics, Getulio Vargas Foundation, Rio de Janeiro 22250, BrazilDepartment of Electrical Engineering, Pontifical Catholic University of Rio de Janeiro, 22451900, BrazilRemote sensing has revolutionized crop mapping and monitoring, providing valuable insights for sustainable agricultural practices. However, successfully implementing remote sensing-based crop type identification in tropical regions remains challenging. Unlike temperate regions, tropical areas benefit from favorable weather conditions that support diverse land management practices, resulting in complex crop dynamics that are difficult to model. Additionally, frequent cloud cover in tropical regions limits the use of optical data during extended periods of the year, making SAR (Synthetic Aperture Radar) a more attractive alternative. Traditional models like Conditional Random Fields (CRFs) have been useful in classifying crop types using spatial and temporal contexts. However, these models often use fixed inputs, optimizing them towards a specific task, and fail to consider the CRFs’ feedback for end-to-end learning features and the spatiotemporal dependencies. This work addresses this gap by introducing an end-to-end hybrid model that combines deep learning with CRFs to incorporate prior knowledge-based modeling of crop dynamics. The proposed framework integrates a backbone encoder–decoder to capture spatial and temporal contexts, with a CRF block that models temporal dynamics by accounting for label dependencies between adjacent epochs. This block facilitates the integration of both data-driven and expert-domain temporal constraints, allowing the framework to adapt to local agricultural practices and enhance crop dynamics modeling. We evaluate the framework using multi-temporal SAR image sequences from two municipalities in Brazil, comparing the effectiveness of learned versus prior-knowledge temporal constraints. The results show improvements of up to 30% in per-class F1 score and 12% in average F1 score compared to a baseline model that excludes temporal dependencies. These findings highlight the value of incorporating prior knowledge-driven temporal constraints into crop mapping models.http://www.sciencedirect.com/science/article/pii/S1569843225002638Crop mappingConditional random fieldsDeep learningHybrid modelMulti-temporal classificationSynthetic aperture radar
spellingShingle Laura Elena Cué La Rosa
Dario Augusto Borges Oliveira
Raul Queiroz Feitosa
Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics
International Journal of Applied Earth Observations and Geoinformation
Crop mapping
Conditional random fields
Deep learning
Hybrid model
Multi-temporal classification
Synthetic aperture radar
title Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics
title_full Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics
title_fullStr Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics
title_full_unstemmed Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics
title_short Harnessing deep learning and CRF for prior-knowledge modeling of crop dynamics
title_sort harnessing deep learning and crf for prior knowledge modeling of crop dynamics
topic Crop mapping
Conditional random fields
Deep learning
Hybrid model
Multi-temporal classification
Synthetic aperture radar
url http://www.sciencedirect.com/science/article/pii/S1569843225002638
work_keys_str_mv AT lauraelenacuelarosa harnessingdeeplearningandcrfforpriorknowledgemodelingofcropdynamics
AT darioaugustoborgesoliveira harnessingdeeplearningandcrfforpriorknowledgemodelingofcropdynamics
AT raulqueirozfeitosa harnessingdeeplearningandcrfforpriorknowledgemodelingofcropdynamics