Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net

Forest ecosystems are critical components of Earth’s biodiversity and play vital roles in climate regulation and carbon sequestration. They face increasing threats from deforestation, wildfires, and other anthropogenic activities. Timely detection and monitoring of changes in forest landscapes pose...

Full description

Saved in:
Bibliographic Details
Main Authors: Ashen Iranga Hewarathna, Luke Hamlin, Joseph Charles, Palanisamy Vigneshwaran, Romiyal George, Selvarajah Thuseethan, Chathrie Wimalasooriya, Bharanidharan Shanmugam
Format: Article
Language:English
Published: MDPI AG 2024-09-01
Series:Technologies
Subjects:
Online Access:https://www.mdpi.com/2227-7080/12/9/160
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850259326863671296
author Ashen Iranga Hewarathna
Luke Hamlin
Joseph Charles
Palanisamy Vigneshwaran
Romiyal George
Selvarajah Thuseethan
Chathrie Wimalasooriya
Bharanidharan Shanmugam
author_facet Ashen Iranga Hewarathna
Luke Hamlin
Joseph Charles
Palanisamy Vigneshwaran
Romiyal George
Selvarajah Thuseethan
Chathrie Wimalasooriya
Bharanidharan Shanmugam
author_sort Ashen Iranga Hewarathna
collection DOAJ
description Forest ecosystems are critical components of Earth’s biodiversity and play vital roles in climate regulation and carbon sequestration. They face increasing threats from deforestation, wildfires, and other anthropogenic activities. Timely detection and monitoring of changes in forest landscapes pose significant challenges for government agencies. To address these challenges, we propose a novel pipeline by refining the U-Net design, including employing two different schemata of early fusion networks and a Siam network architecture capable of processing RGB images specifically designed to identify high-risk areas in forest ecosystems through change detection across different time frames in the same location. It annotates ground truth change maps in such time frames using an encoder–decoder approach with the help of an enhanced feature learning and attention mechanism. Our proposed pipeline, integrated with ResNeSt blocks and SE attention techniques, achieved impressive results in our newly created forest cover change dataset. The evaluation metrics reveal a Dice score of 39.03%, a kappa score of 35.13%, an F1-score of 42.84%, and an overall accuracy of 94.37%. Notably, our approach significantly outperformed multitasking model approaches in the ONERA dataset, boasting a precision of 53.32%, a Dice score of 59.97%, and an overall accuracy of 97.82%. Furthermore, it surpassed multitasking models in the HRSCD dataset, even without utilizing land cover maps, achieving a Dice score of 44.62%, a kappa score of 11.97%, and an overall accuracy of 98.44%. Although the proposed model had a lower F1-score than other methods, other performance metrics highlight its effectiveness in timely detection and forest landscape monitoring, advancing deep learning techniques in this field.
format Article
id doaj-art-72fd5dbd67c3490f91f14031d94bfd44
institution OA Journals
issn 2227-7080
language English
publishDate 2024-09-01
publisher MDPI AG
record_format Article
series Technologies
spelling doaj-art-72fd5dbd67c3490f91f14031d94bfd442025-08-20T01:55:52ZengMDPI AGTechnologies2227-70802024-09-0112916010.3390/technologies12090160Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-NetAshen Iranga Hewarathna0Luke Hamlin1Joseph Charles2Palanisamy Vigneshwaran3Romiyal George4Selvarajah Thuseethan5Chathrie Wimalasooriya6Bharanidharan Shanmugam7Faculty of Computing, Sabaragamuwa University of Sri Lanka, Belihuloya 70140, Sri LankaEnergy and Resource Institute, Faculty of Science and Technology, Charles Darwin University, Darwin, NT 0810, AustraliaFaculty of Engineering, Friedrich-Alexander-University (FAU), 91054 Erlangen, GermanyFaculty of Computing, Sabaragamuwa University of Sri Lanka, Belihuloya 70140, Sri LankaDepartment of Computer Engineering, University of Peradeniya, Peradeniya 20400, Sri LankaEnergy and Resource Institute, Faculty of Science and Technology, Charles Darwin University, Darwin, NT 0810, AustraliaSchool of Computing, University of Otago, Dunedin 9016, New ZealandEnergy and Resource Institute, Faculty of Science and Technology, Charles Darwin University, Darwin, NT 0810, AustraliaForest ecosystems are critical components of Earth’s biodiversity and play vital roles in climate regulation and carbon sequestration. They face increasing threats from deforestation, wildfires, and other anthropogenic activities. Timely detection and monitoring of changes in forest landscapes pose significant challenges for government agencies. To address these challenges, we propose a novel pipeline by refining the U-Net design, including employing two different schemata of early fusion networks and a Siam network architecture capable of processing RGB images specifically designed to identify high-risk areas in forest ecosystems through change detection across different time frames in the same location. It annotates ground truth change maps in such time frames using an encoder–decoder approach with the help of an enhanced feature learning and attention mechanism. Our proposed pipeline, integrated with ResNeSt blocks and SE attention techniques, achieved impressive results in our newly created forest cover change dataset. The evaluation metrics reveal a Dice score of 39.03%, a kappa score of 35.13%, an F1-score of 42.84%, and an overall accuracy of 94.37%. Notably, our approach significantly outperformed multitasking model approaches in the ONERA dataset, boasting a precision of 53.32%, a Dice score of 59.97%, and an overall accuracy of 97.82%. Furthermore, it surpassed multitasking models in the HRSCD dataset, even without utilizing land cover maps, achieving a Dice score of 44.62%, a kappa score of 11.97%, and an overall accuracy of 98.44%. Although the proposed model had a lower F1-score than other methods, other performance metrics highlight its effectiveness in timely detection and forest landscape monitoring, advancing deep learning techniques in this field.https://www.mdpi.com/2227-7080/12/9/160deforestationchange detectionSiamese attention mechanismU-Net
spellingShingle Ashen Iranga Hewarathna
Luke Hamlin
Joseph Charles
Palanisamy Vigneshwaran
Romiyal George
Selvarajah Thuseethan
Chathrie Wimalasooriya
Bharanidharan Shanmugam
Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
Technologies
deforestation
change detection
Siamese attention mechanism
U-Net
title Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
title_full Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
title_fullStr Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
title_full_unstemmed Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
title_short Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
title_sort change detection for forest ecosystems using remote sensing images with siamese attention u net
topic deforestation
change detection
Siamese attention mechanism
U-Net
url https://www.mdpi.com/2227-7080/12/9/160
work_keys_str_mv AT ashenirangahewarathna changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT lukehamlin changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT josephcharles changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT palanisamyvigneshwaran changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT romiyalgeorge changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT selvarajahthuseethan changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT chathriewimalasooriya changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet
AT bharanidharanshanmugam changedetectionforforestecosystemsusingremotesensingimageswithsiameseattentionunet