A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance
Mesoscale eddies are pivotal oceanographic phenomena affecting marine environments. Accurate and stable identification of these eddies is essential for advancing research on their dynamics and effects. Current methods primarily focus on identifying Cyclonic and Anticyclonic eddies (CE, AE), with ano...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/2/457 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587490212446208 |
|---|---|
| author | Lei Zhang Xiaodong Ma Weishuai Xu Xiang Wan Qiyun Chen |
| author_facet | Lei Zhang Xiaodong Ma Weishuai Xu Xiang Wan Qiyun Chen |
| author_sort | Lei Zhang |
| collection | DOAJ |
| description | Mesoscale eddies are pivotal oceanographic phenomena affecting marine environments. Accurate and stable identification of these eddies is essential for advancing research on their dynamics and effects. Current methods primarily focus on identifying Cyclonic and Anticyclonic eddies (CE, AE), with anomalous eddy identification often requiring secondary analyses of sea surface height anomalies and eddy center properties, leading to segmented data interpretations. This study introduces a deep learning model integrating multi-source fusion data with a Squeeze-and-Excitation (SE) attention mechanism to enhance the identification accuracy for both normal and anomalous eddies. Comparative ablation experiments validate the model’s effectiveness, demonstrating its potential for more nuanced, multi-source, and multi-class mesoscale eddy identification. This approach offers a promising framework for advancing mesoscale eddy identification through deep learning. |
| format | Article |
| id | doaj-art-4298c1d3acec46dabe670d29828f33e3 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-4298c1d3acec46dabe670d29828f33e32025-01-24T13:49:00ZengMDPI AGSensors1424-82202025-01-0125245710.3390/s25020457A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori GuidanceLei Zhang0Xiaodong Ma1Weishuai Xu2Xiang Wan3Qiyun Chen4Department of Military and Marine Mapping, Dalian Naval Academy, Dalian 116021, ChinaDalian Naval Academy Cadet Brigade, Dalian 116000, ChinaDalian Naval Academy Cadet Brigade, Dalian 116000, ChinaDalian Naval Academy Cadet Brigade, Dalian 116000, ChinaDalian Naval Academy Cadet Brigade, Dalian 116000, ChinaMesoscale eddies are pivotal oceanographic phenomena affecting marine environments. Accurate and stable identification of these eddies is essential for advancing research on their dynamics and effects. Current methods primarily focus on identifying Cyclonic and Anticyclonic eddies (CE, AE), with anomalous eddy identification often requiring secondary analyses of sea surface height anomalies and eddy center properties, leading to segmented data interpretations. This study introduces a deep learning model integrating multi-source fusion data with a Squeeze-and-Excitation (SE) attention mechanism to enhance the identification accuracy for both normal and anomalous eddies. Comparative ablation experiments validate the model’s effectiveness, demonstrating its potential for more nuanced, multi-source, and multi-class mesoscale eddy identification. This approach offers a promising framework for advancing mesoscale eddy identification through deep learning.https://www.mdpi.com/1424-8220/25/2/457mesoscale eddyidentification methoddeep learningYOLO |
| spellingShingle | Lei Zhang Xiaodong Ma Weishuai Xu Xiang Wan Qiyun Chen A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance Sensors mesoscale eddy identification method deep learning YOLO |
| title | A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance |
| title_full | A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance |
| title_fullStr | A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance |
| title_full_unstemmed | A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance |
| title_short | A Novel Mesoscale Eddy Identification Method Using Enhanced Interpolation and A Posteriori Guidance |
| title_sort | novel mesoscale eddy identification method using enhanced interpolation and a posteriori guidance |
| topic | mesoscale eddy identification method deep learning YOLO |
| url | https://www.mdpi.com/1424-8220/25/2/457 |
| work_keys_str_mv | AT leizhang anovelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT xiaodongma anovelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT weishuaixu anovelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT xiangwan anovelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT qiyunchen anovelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT leizhang novelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT xiaodongma novelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT weishuaixu novelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT xiangwan novelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance AT qiyunchen novelmesoscaleeddyidentificationmethodusingenhancedinterpolationandaposterioriguidance |