SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer
Mangrove wetlands play a crucial role in maintaining species diversity. However, they face threats from habitat degradation, deforestation, pollution, and climate change. Detecting changes in mangrove wetlands is essential for understanding their ecological implications, but it remains a challenging...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2024-10-01
|
| Series: | Electronic Research Archive |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/era.2024260 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832590743633395712 |
|---|---|
| author | Zhenhua Wang Jinlong Yang Chuansheng Dong Xi Zhang Congqin Yi Jiuhu Sun |
| author_facet | Zhenhua Wang Jinlong Yang Chuansheng Dong Xi Zhang Congqin Yi Jiuhu Sun |
| author_sort | Zhenhua Wang |
| collection | DOAJ |
| description | Mangrove wetlands play a crucial role in maintaining species diversity. However, they face threats from habitat degradation, deforestation, pollution, and climate change. Detecting changes in mangrove wetlands is essential for understanding their ecological implications, but it remains a challenging task. In this study, we propose a semantic segmentation model for mangroves based on Deeplabv3+ with Swin Transformer, abbreviated as SSMM-DS. Using Deeplabv3+ as the basic framework, we first constructed a data concatenation module to improve the contrast between mangroves and other vegetation or water. We then employed Swin Transformer as the backbone network, enhancing the capability of global information learning and detail feature extraction. Finally, we optimized the loss function by combining cross-entropy loss and dice loss, addressing the issue of sampling imbalance caused by the small areas of mangroves. Using GF-1 and GF-6 images, taking mean precision (mPrecision), mean intersection over union (mIoU), floating-point operations (FLOPs), and the number of parameters (Params) as evaluation metrics, we evaluate SSMM-DS against state-of-the-art models, including FCN, PSPNet, OCRNet, uPerNet, and SegFormer. The results demonstrate SSMM-DS's superiority in terms of mIoU, mPrecision, and parameter efficiency. SSMM-DS achieves a higher mIoU (95.11%) and mPrecision (97.79%) while using fewer parameters (17.48M) compared to others. Although its FLOPs are slightly higher than SegFormer's (15.11G vs. 9.9G), SSMM-DS offers a balance between performance and efficiency. Experimental results highlight SSMM-DS's effectiveness in extracting mangrove features, making it a valuable tool for monitoring and managing these critical ecosystems. |
| format | Article |
| id | doaj-art-56c0bd0978ac4ffe9526ac51a737a6f3 |
| institution | Kabale University |
| issn | 2688-1594 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | Electronic Research Archive |
| spelling | doaj-art-56c0bd0978ac4ffe9526ac51a737a6f32025-01-23T07:52:52ZengAIMS PressElectronic Research Archive2688-15942024-10-0132105615563210.3934/era.2024260SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformerZhenhua Wang0Jinlong Yang1Chuansheng Dong2Xi Zhang3Congqin Yi4Jiuhu Sun5College of Information Technology, Shanghai Ocean University, Shanghai 201306, ChinaCollege of Information Technology, Shanghai Ocean University, Shanghai 201306, ChinaShandong Land Surveying and Mapping Institute, Jinan 250000, ChinaShandong provincial institute of land space data and remote sensing technology, Jinan 250000, ChinaCollege of Information Technology, Shanghai Ocean University, Shanghai 201306, ChinaShandong Land Surveying and Mapping Institute, Jinan 250000, ChinaMangrove wetlands play a crucial role in maintaining species diversity. However, they face threats from habitat degradation, deforestation, pollution, and climate change. Detecting changes in mangrove wetlands is essential for understanding their ecological implications, but it remains a challenging task. In this study, we propose a semantic segmentation model for mangroves based on Deeplabv3+ with Swin Transformer, abbreviated as SSMM-DS. Using Deeplabv3+ as the basic framework, we first constructed a data concatenation module to improve the contrast between mangroves and other vegetation or water. We then employed Swin Transformer as the backbone network, enhancing the capability of global information learning and detail feature extraction. Finally, we optimized the loss function by combining cross-entropy loss and dice loss, addressing the issue of sampling imbalance caused by the small areas of mangroves. Using GF-1 and GF-6 images, taking mean precision (mPrecision), mean intersection over union (mIoU), floating-point operations (FLOPs), and the number of parameters (Params) as evaluation metrics, we evaluate SSMM-DS against state-of-the-art models, including FCN, PSPNet, OCRNet, uPerNet, and SegFormer. The results demonstrate SSMM-DS's superiority in terms of mIoU, mPrecision, and parameter efficiency. SSMM-DS achieves a higher mIoU (95.11%) and mPrecision (97.79%) while using fewer parameters (17.48M) compared to others. Although its FLOPs are slightly higher than SegFormer's (15.11G vs. 9.9G), SSMM-DS offers a balance between performance and efficiency. Experimental results highlight SSMM-DS's effectiveness in extracting mangrove features, making it a valuable tool for monitoring and managing these critical ecosystems.https://www.aimspress.com/article/doi/10.3934/era.2024260remote sensing imagesemantic segmentationmangroveswin transformer |
| spellingShingle | Zhenhua Wang Jinlong Yang Chuansheng Dong Xi Zhang Congqin Yi Jiuhu Sun SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer Electronic Research Archive remote sensing image semantic segmentation mangrove swin transformer |
| title | SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer |
| title_full | SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer |
| title_fullStr | SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer |
| title_full_unstemmed | SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer |
| title_short | SSMM-DS: A semantic segmentation model for mangroves based on Deeplabv3+ with swin transformer |
| title_sort | ssmm ds a semantic segmentation model for mangroves based on deeplabv3 with swin transformer |
| topic | remote sensing image semantic segmentation mangrove swin transformer |
| url | https://www.aimspress.com/article/doi/10.3934/era.2024260 |
| work_keys_str_mv | AT zhenhuawang ssmmdsasemanticsegmentationmodelformangrovesbasedondeeplabv3withswintransformer AT jinlongyang ssmmdsasemanticsegmentationmodelformangrovesbasedondeeplabv3withswintransformer AT chuanshengdong ssmmdsasemanticsegmentationmodelformangrovesbasedondeeplabv3withswintransformer AT xizhang ssmmdsasemanticsegmentationmodelformangrovesbasedondeeplabv3withswintransformer AT congqinyi ssmmdsasemanticsegmentationmodelformangrovesbasedondeeplabv3withswintransformer AT jiuhusun ssmmdsasemanticsegmentationmodelformangrovesbasedondeeplabv3withswintransformer |