Global River Monitoring Using Semantic Fusion Networks
Global river monitoring is an important mission within the remote sensing society. One of the main challenges faced by this mission is generating an accurate water mask from remote sensing images (RSI) of rivers (RSIR), especially on a global scale with various river features. Aiming at better water...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Water |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4441/12/8/2258 |
| _version_ | 1797558817869791232 |
|---|---|
| author | Zhihao Wei Kebin Jia Xiaowei Jia Ankush Khandelwal Vipin Kumar |
| author_facet | Zhihao Wei Kebin Jia Xiaowei Jia Ankush Khandelwal Vipin Kumar |
| author_sort | Zhihao Wei |
| collection | DOAJ |
| description | Global river monitoring is an important mission within the remote sensing society. One of the main challenges faced by this mission is generating an accurate water mask from remote sensing images (RSI) of rivers (RSIR), especially on a global scale with various river features. Aiming at better water area classification using semantic information, this paper presents a segmentation method for global river monitoring based on semantic clustering and semantic fusion. Firstly, an encoder–decoder network (AEN)-based architecture is proposed to obtain the semantic features from RSIR. Secondly, a clustering-based semantic fusion method is proposed to divide semantic features of RSIR into groups and train convolutional neural networks (CNN) models corresponding to each group using data augmentation and semi-supervised learning. Thirdly, a semantic distance-based segmentation fusion method is proposed for fusing the CNN models result into final segmentation mask. We built a global river dataset that contains multiple river segments from each continent of the world based on Sentinel-2 satellite imagery. The result shows that the F1-score of the proposed segmentation method is 93.32%, which outperforms several state-of-the-art algorithms, and demonstrates that grouping semantic information helps better segment the RSIR in global scale. |
| first_indexed | 2024-03-10T17:35:45Z |
| format | Article |
| id | doaj.art-aa6b1c1340c84a8c8ab138f41a85d221 |
| institution | Directory Open Access Journal |
| issn | 2073-4441 |
| language | English |
| last_indexed | 2024-03-10T17:35:45Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Water |
| spelling | doaj.art-aa6b1c1340c84a8c8ab138f41a85d2212023-11-20T09:51:27ZengMDPI AGWater2073-44412020-08-01128225810.3390/w12082258Global River Monitoring Using Semantic Fusion NetworksZhihao Wei0Kebin Jia1Xiaowei Jia2Ankush Khandelwal3Vipin Kumar4Department of Information and Communication Engineering, Faculty of Information Technology, Beijing University of Technology, Beijing 100124, ChinaDepartment of Information and Communication Engineering, Faculty of Information Technology, Beijing University of Technology, Beijing 100124, ChinaDepartment of Computer Science and Engineering, University of Minnesota, Minneapolis, MN 55455, USADepartment of Computer Science and Engineering, University of Minnesota, Minneapolis, MN 55455, USADepartment of Computer Science and Engineering, University of Minnesota, Minneapolis, MN 55455, USAGlobal river monitoring is an important mission within the remote sensing society. One of the main challenges faced by this mission is generating an accurate water mask from remote sensing images (RSI) of rivers (RSIR), especially on a global scale with various river features. Aiming at better water area classification using semantic information, this paper presents a segmentation method for global river monitoring based on semantic clustering and semantic fusion. Firstly, an encoder–decoder network (AEN)-based architecture is proposed to obtain the semantic features from RSIR. Secondly, a clustering-based semantic fusion method is proposed to divide semantic features of RSIR into groups and train convolutional neural networks (CNN) models corresponding to each group using data augmentation and semi-supervised learning. Thirdly, a semantic distance-based segmentation fusion method is proposed for fusing the CNN models result into final segmentation mask. We built a global river dataset that contains multiple river segments from each continent of the world based on Sentinel-2 satellite imagery. The result shows that the F1-score of the proposed segmentation method is 93.32%, which outperforms several state-of-the-art algorithms, and demonstrates that grouping semantic information helps better segment the RSIR in global scale.https://www.mdpi.com/2073-4441/12/8/2258convolutionencoder–decoder networkfeature extractionremote sensing image of riversemantic fusionsemi-supervised learning |
| spellingShingle | Zhihao Wei Kebin Jia Xiaowei Jia Ankush Khandelwal Vipin Kumar Global River Monitoring Using Semantic Fusion Networks Water convolution encoder–decoder network feature extraction remote sensing image of river semantic fusion semi-supervised learning |
| title | Global River Monitoring Using Semantic Fusion Networks |
| title_full | Global River Monitoring Using Semantic Fusion Networks |
| title_fullStr | Global River Monitoring Using Semantic Fusion Networks |
| title_full_unstemmed | Global River Monitoring Using Semantic Fusion Networks |
| title_short | Global River Monitoring Using Semantic Fusion Networks |
| title_sort | global river monitoring using semantic fusion networks |
| topic | convolution encoder–decoder network feature extraction remote sensing image of river semantic fusion semi-supervised learning |
| url | https://www.mdpi.com/2073-4441/12/8/2258 |
| work_keys_str_mv | AT zhihaowei globalrivermonitoringusingsemanticfusionnetworks AT kebinjia globalrivermonitoringusingsemanticfusionnetworks AT xiaoweijia globalrivermonitoringusingsemanticfusionnetworks AT ankushkhandelwal globalrivermonitoringusingsemanticfusionnetworks AT vipinkumar globalrivermonitoringusingsemanticfusionnetworks |