Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning
Accurate bathymetric data are critical for marine ecological balance and resource management. Deep learning algorithms, known for their capacity to model complex, multivariate, and nonlinear relationships, have been increasingly applied to satellite-derived bathymetry. However, existing deep learnin...
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/11080361/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849240466789236736 |
|---|---|
| author | Weidong Zhu Yanying Huang Tiantian Cao Xiaoshan Zhang Qidi Xie Kuifeng Luan Wei Shen Ziya Zou |
| author_facet | Weidong Zhu Yanying Huang Tiantian Cao Xiaoshan Zhang Qidi Xie Kuifeng Luan Wei Shen Ziya Zou |
| author_sort | Weidong Zhu |
| collection | DOAJ |
| description | Accurate bathymetric data are critical for marine ecological balance and resource management. Deep learning algorithms, known for their capacity to model complex, multivariate, and nonlinear relationships, have been increasingly applied to satellite-derived bathymetry. However, existing deep learning models are limited by simple architectures and low efficiency in hyperparameter optimization, resulting in suboptimal training performance. This article proposes a convolutional neural network and bidirectional long short-term memory hybrid model based on the Bayesian optimization algorithm (BOA-CNN-BILSTM) to enhance bathymetric inversion accuracy and efficiency. The model employs BOA to optimize the key hyperparameters of the CNN-BILSTM architecture, thereby improving inversion performance. Bathymetric inversion experiments were conducted using fused ICESat-2 and Sentinel-2 data, focusing on Coral Island and Dong Island in the South China Sea, as well as Midway Island and Oahu Island in the Pacific Ocean. Comparative experiments demonstrated that BOA significantly outperforms conventional random search by achieving near-optimal hyperparameter configurations with fewer evaluations, accelerating convergence and reducing computational costs. The BOA-CNN-BILSTM model reduced the root mean square error by 28.6%–56.5%, 29.6%–53.7%, 34.1%–52.6%, and 28.9%–57.1% across the study areas compared with the CNN, BILSTM, and CNN-BILSTM models. Other evaluation metrics also showed varying degrees of improvement. These results demonstrate that the proposed approach is effective and highly accurate for bathymetric inversion in shallow waters. |
| format | Article |
| id | doaj-art-bb47251a16154e7d93cb7ea55fcf547d |
| institution | Kabale University |
| 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-bb47251a16154e7d93cb7ea55fcf547d2025-08-20T04:00:34ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118183761839010.1109/JSTARS.2025.358928911080361Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep LearningWeidong Zhu0https://orcid.org/0000-0002-6404-0891Yanying Huang1https://orcid.org/0009-0009-5103-4538Tiantian Cao2https://orcid.org/0009-0001-2737-6177Xiaoshan Zhang3Qidi Xie4Kuifeng Luan5https://orcid.org/0000-0001-7009-3888Wei Shen6https://orcid.org/0009-0003-5857-3899Ziya Zou7https://orcid.org/0000-0002-6916-1078School of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaSchool of Oceanography and Ecological Science, Shanghai Ocean University, Shanghai, ChinaAccurate bathymetric data are critical for marine ecological balance and resource management. Deep learning algorithms, known for their capacity to model complex, multivariate, and nonlinear relationships, have been increasingly applied to satellite-derived bathymetry. However, existing deep learning models are limited by simple architectures and low efficiency in hyperparameter optimization, resulting in suboptimal training performance. This article proposes a convolutional neural network and bidirectional long short-term memory hybrid model based on the Bayesian optimization algorithm (BOA-CNN-BILSTM) to enhance bathymetric inversion accuracy and efficiency. The model employs BOA to optimize the key hyperparameters of the CNN-BILSTM architecture, thereby improving inversion performance. Bathymetric inversion experiments were conducted using fused ICESat-2 and Sentinel-2 data, focusing on Coral Island and Dong Island in the South China Sea, as well as Midway Island and Oahu Island in the Pacific Ocean. Comparative experiments demonstrated that BOA significantly outperforms conventional random search by achieving near-optimal hyperparameter configurations with fewer evaluations, accelerating convergence and reducing computational costs. The BOA-CNN-BILSTM model reduced the root mean square error by 28.6%–56.5%, 29.6%–53.7%, 34.1%–52.6%, and 28.9%–57.1% across the study areas compared with the CNN, BILSTM, and CNN-BILSTM models. Other evaluation metrics also showed varying degrees of improvement. These results demonstrate that the proposed approach is effective and highly accurate for bathymetric inversion in shallow waters.https://ieeexplore.ieee.org/document/11080361/Deep learning modelICESat-2satellite-derived bathymetry (SDB)Sentinel-2 |
| spellingShingle | Weidong Zhu Yanying Huang Tiantian Cao Xiaoshan Zhang Qidi Xie Kuifeng Luan Wei Shen Ziya Zou Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning model ICESat-2 satellite-derived bathymetry (SDB) Sentinel-2 |
| title | Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning |
| title_full | Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning |
| title_fullStr | Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning |
| title_full_unstemmed | Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning |
| title_short | Satellite-Derived Bathymetry Combined With Sentinel-2 and ICESat-2 Datasets Using Deep Learning |
| title_sort | satellite derived bathymetry combined with sentinel 2 and icesat 2 datasets using deep learning |
| topic | Deep learning model ICESat-2 satellite-derived bathymetry (SDB) Sentinel-2 |
| url | https://ieeexplore.ieee.org/document/11080361/ |
| work_keys_str_mv | AT weidongzhu satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT yanyinghuang satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT tiantiancao satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT xiaoshanzhang satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT qidixie satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT kuifengluan satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT weishen satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning AT ziyazou satellitederivedbathymetrycombinedwithsentinel2andicesat2datasetsusingdeeplearning |