Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications
Monitoring glacial lakes is critical for assessing climate change impacts and mitigating glacial lake outburst flood risks. This study evaluates three deep learning architectures, U-Net, simple convolutional neural network (CNN), and atrous spatial pyramid pooling SegNet (ASPP SegNet), for binary se...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| 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/11021621/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839648905781837824 |
|---|---|
| author | Lingling Xue Asad Khan Muhammad Haseeb Mourad Aqnouy Dawood Ahmad Refka Ghodhbani Dmitry E. Kucher Olga D. Kucher |
| author_facet | Lingling Xue Asad Khan Muhammad Haseeb Mourad Aqnouy Dawood Ahmad Refka Ghodhbani Dmitry E. Kucher Olga D. Kucher |
| author_sort | Lingling Xue |
| collection | DOAJ |
| description | Monitoring glacial lakes is critical for assessing climate change impacts and mitigating glacial lake outburst flood risks. This study evaluates three deep learning architectures, U-Net, simple convolutional neural network (CNN), and atrous spatial pyramid pooling SegNet (ASPP SegNet), for binary semantic segmentation of glacial lakes using multisensor optical satellite imagery (Sentinel-2). Incorporating data augmentation and custom evaluation metrics (IoU, F1-score, validation loss), the results show that Simple CNN achieves the highest IoU (0.9155) and F1-score (0.9557). At the same time, ASPP SegNet demonstrates superior generalization with the lowest validation loss (0.03337). U-Net also delivers a reliable performance, albeit slightly lower. Visual and quantitative assessments highlight the advantage of multiscale, context-aware architectures in delineating fragmented lake boundaries. This comparative study provides practical guidance for deep learning model selection in remote sensing-based glacial and coastal hydrology monitoring. Future work will explore temporal modeling, multiclass segmentation, and the integration of optical, radar, and elevation data for improved resilience. |
| format | Article |
| id | doaj-art-bee33f2a738b48d5a151b26d29e4a7df |
| institution | Matheson Library |
| 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-bee33f2a738b48d5a151b26d29e4a7df2025-06-27T23:00:23ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118153041531710.1109/JSTARS.2025.357618711021621Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology ApplicationsLingling Xue0Asad Khan1https://orcid.org/0000-0002-1261-0418Muhammad Haseeb2https://orcid.org/0000-0002-3473-6396Mourad Aqnouy3https://orcid.org/0000-0003-1633-2870Dawood Ahmad4Refka Ghodhbani5Dmitry E. Kucher6https://orcid.org/0000-0002-7919-3487Olga D. Kucher7https://orcid.org/0009-0002-9816-7759College of Energy Engineering, Huanghuai University, Zhumadian, ChinaMetaverse Research Institute, School of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou, ChinaInstitute of Space Sciences, University of the Punjab, Lahore, PakistanLaboratory of Applied and Marine Geosciences, Geotechnics and Geohazards (LR3G), Faculty of Sciences, Abdelmalek Essaadi University, Tetouan, MoroccoInstitute of Space Sciences, University of the Punjab, Lahore, PakistanCenter for Scientific Research and Entrepreneurship, Northern Border University, Arar, Saudi ArabiaDepartment of Environmental Management, Institute of Environmental Engineering, RUDN University, Moscow, RussiaDepartment of Environmental Management, Institute of Environmental Engineering, RUDN University, Moscow, RussiaMonitoring glacial lakes is critical for assessing climate change impacts and mitigating glacial lake outburst flood risks. This study evaluates three deep learning architectures, U-Net, simple convolutional neural network (CNN), and atrous spatial pyramid pooling SegNet (ASPP SegNet), for binary semantic segmentation of glacial lakes using multisensor optical satellite imagery (Sentinel-2). Incorporating data augmentation and custom evaluation metrics (IoU, F1-score, validation loss), the results show that Simple CNN achieves the highest IoU (0.9155) and F1-score (0.9557). At the same time, ASPP SegNet demonstrates superior generalization with the lowest validation loss (0.03337). U-Net also delivers a reliable performance, albeit slightly lower. Visual and quantitative assessments highlight the advantage of multiscale, context-aware architectures in delineating fragmented lake boundaries. This comparative study provides practical guidance for deep learning model selection in remote sensing-based glacial and coastal hydrology monitoring. Future work will explore temporal modeling, multiclass segmentation, and the integration of optical, radar, and elevation data for improved resilience.https://ieeexplore.ieee.org/document/11021621/Atrous spatial pyramid pooling SegNet (ASPP SegNet)coastal monitoringdeep learningglacial lakeremote sensingsemantic segmentation |
| spellingShingle | Lingling Xue Asad Khan Muhammad Haseeb Mourad Aqnouy Dawood Ahmad Refka Ghodhbani Dmitry E. Kucher Olga D. Kucher Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Atrous spatial pyramid pooling SegNet (ASPP SegNet) coastal monitoring deep learning glacial lake remote sensing semantic segmentation |
| title | Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications |
| title_full | Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications |
| title_fullStr | Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications |
| title_full_unstemmed | Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications |
| title_short | Multisensor Fusion and Deep Learning Approaches for Semantic Segmentation of Glacial Lakes: A Comparative Study for Coastal Hydrology Applications |
| title_sort | multisensor fusion and deep learning approaches for semantic segmentation of glacial lakes a comparative study for coastal hydrology applications |
| topic | Atrous spatial pyramid pooling SegNet (ASPP SegNet) coastal monitoring deep learning glacial lake remote sensing semantic segmentation |
| url | https://ieeexplore.ieee.org/document/11021621/ |
| work_keys_str_mv | AT linglingxue multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT asadkhan multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT muhammadhaseeb multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT mouradaqnouy multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT dawoodahmad multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT refkaghodhbani multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT dmitryekucher multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications AT olgadkucher multisensorfusionanddeeplearningapproachesforsemanticsegmentationofglaciallakesacomparativestudyforcoastalhydrologyapplications |