Change Detection for Forest Ecosystems Using Remote Sensing Images with Siamese Attention U-Net
Authors:
- Ashen Iranga HewarathnaAHAshen Iranga Hewarathna
- Luke HamlinLHLuke Hamlin
- Joseph CharlesJCJoseph Charles
- Palanisamy VigneshwaranPVPalanisamy Vigneshwaran
- Romiyal GeorgeRGRomiyal George
- Selvarajah ThuseethanSTSelvarajah Thuseethan
- Chathrie WimalasooriyaCWChathrie Wimalasooriya
- Bharanidharan ShanmugamBSBharanidharan Shanmugam
Open Access
- 12 September 2024
- journal article
- research article
- Published by MDPI AG in Technologies
- Vol. 12 (9) , 160
Abstract
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.Keywords
Funding Information
- Charles Darwin University
This publication has 39 references indexed in Scilit:
- Advances in the study of global forest wildfiresJournal of Soils and Sediments, 2023
- It's all in the mud - The use of sediment geochemistry to estimate contemporary water quality in lakesApplied Geochemistry, 2023
- Management linked to ecological intensification supports insect pollinators in Iberian wood-pasturesLandscape Ecology, 2023
- Spatiotemporal dynamics of forest insect populations under climate changeCurrent Opinion in Insect Science, 2023
- Predicting muscle invasion in bladder cancer based on MRI: A comparison of radiomics, and single-task and multi-task deep learningComputer Methods and Programs in Biomedicine, 2023
- Construction of ecological security pattern of national ecological barriers for ecosystem health maintenanceEcological Indicators, 2022
- Mega‐disturbances cause rapid decline of mature conifer forest habitat in CaliforniaEcological Applications, 2022
- Climate warming exacerbates plant disease through enhancing commensal interaction of co-infested insect vectorsJournal of Pest Science, 2022
- Focus U-Net: A novel dual attention-gated CNN for polyp segmentation during colonoscopyComputers in Biology and Medicine, 2021
- Multitask learning for large-scale semantic change detectionComputer Vision and Image Understanding, 2019