Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets
Identifying permanent water and temporary water in flood disasters efficiently has mainly relied on change detection method from multi-temporal remote sensing imageries, but estimating the water type in flood disaster events from only post-flood remote sensing imageries still remains challenging. Re...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/11/2220 |
| _version_ | 1797531252182482944 |
|---|---|
| author | Yanbing Bai Wenqi Wu Zhengxin Yang Jinze Yu Bo Zhao Xing Liu Hanfang Yang Erick Mas Shunichi Koshimura |
| author_facet | Yanbing Bai Wenqi Wu Zhengxin Yang Jinze Yu Bo Zhao Xing Liu Hanfang Yang Erick Mas Shunichi Koshimura |
| author_sort | Yanbing Bai |
| collection | DOAJ |
| description | Identifying permanent water and temporary water in flood disasters efficiently has mainly relied on change detection method from multi-temporal remote sensing imageries, but estimating the water type in flood disaster events from only post-flood remote sensing imageries still remains challenging. Research progress in recent years has demonstrated the excellent potential of multi-source data fusion and deep learning algorithms in improving flood detection, while this field has only been studied initially due to the lack of large-scale labelled remote sensing images of flood events. Here, we present new deep learning algorithms and a multi-source data fusion driven flood inundation mapping approach by leveraging a large-scale publicly available Sen1Flood11 dataset consisting of roughly 4831 labelled Sentinel-1 SAR and Sentinel-2 optical imagery gathered from flood events worldwide in recent years. Specifically, we proposed an automatic segmentation method for surface water, permanent water, and temporary water identification, and all tasks share the same convolutional neural network architecture. We utilize focal loss to deal with the class (water/non-water) imbalance problem. Thorough ablation experiments and analysis confirmed the effectiveness of various proposed designs. In comparison experiments, the method proposed in this paper is superior to other classical models. Our model achieves a mean Intersection over Union (mIoU) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>52.99</mn><mo>%</mo></mrow></semantics></math></inline-formula>, Intersection over Union (IoU) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>52.30</mn><mo>%</mo></mrow></semantics></math></inline-formula>, and Overall Accuracy (OA) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>92.81</mn><mo>%</mo></mrow></semantics></math></inline-formula> on the Sen1Flood11 test set. On the Sen1Flood11 Bolivia test set, our model also achieves very high mIoU (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>47.88</mn><mo>%</mo></mrow></semantics></math></inline-formula>), IoU (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>76.74</mn><mo>%</mo></mrow></semantics></math></inline-formula>), and OA (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>95.59</mn><mo>%</mo></mrow></semantics></math></inline-formula>) and shows good generalization ability. |
| first_indexed | 2024-03-10T10:41:12Z |
| format | Article |
| id | doaj.art-9fd78fd0687a4424b4fca82c4c65da33 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T10:41:12Z |
| publishDate | 2021-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-9fd78fd0687a4424b4fca82c4c65da332023-11-21T22:57:55ZengMDPI AGRemote Sensing2072-42922021-06-011311222010.3390/rs13112220Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark DatasetsYanbing Bai0Wenqi Wu1Zhengxin Yang2Jinze Yu3Bo Zhao4Xing Liu5Hanfang Yang6Erick Mas7Shunichi Koshimura8Center for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaCenter for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaChina Huaneng Group Co., Ltd., Beijing 100031, ChinaGraduate School of Information Science and Technology, The University of Tokyo, Tokyo 113-8656, JapanSchool of Informatics, The University of Edinburgh, Edinburgh EH8 9AB, UKGraduate School of Information Sciences, Tohoku University, Sendai 980-8579, JapanCenter for Applied Statistics, School of Statistics, Renmin University of China, Beijing 100872, ChinaInternational Research Institute of Disaster Science, Tohoku University, Sendai 980-8572, JapanInternational Research Institute of Disaster Science, Tohoku University, Sendai 980-8572, JapanIdentifying permanent water and temporary water in flood disasters efficiently has mainly relied on change detection method from multi-temporal remote sensing imageries, but estimating the water type in flood disaster events from only post-flood remote sensing imageries still remains challenging. Research progress in recent years has demonstrated the excellent potential of multi-source data fusion and deep learning algorithms in improving flood detection, while this field has only been studied initially due to the lack of large-scale labelled remote sensing images of flood events. Here, we present new deep learning algorithms and a multi-source data fusion driven flood inundation mapping approach by leveraging a large-scale publicly available Sen1Flood11 dataset consisting of roughly 4831 labelled Sentinel-1 SAR and Sentinel-2 optical imagery gathered from flood events worldwide in recent years. Specifically, we proposed an automatic segmentation method for surface water, permanent water, and temporary water identification, and all tasks share the same convolutional neural network architecture. We utilize focal loss to deal with the class (water/non-water) imbalance problem. Thorough ablation experiments and analysis confirmed the effectiveness of various proposed designs. In comparison experiments, the method proposed in this paper is superior to other classical models. Our model achieves a mean Intersection over Union (mIoU) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>52.99</mn><mo>%</mo></mrow></semantics></math></inline-formula>, Intersection over Union (IoU) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>52.30</mn><mo>%</mo></mrow></semantics></math></inline-formula>, and Overall Accuracy (OA) of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>92.81</mn><mo>%</mo></mrow></semantics></math></inline-formula> on the Sen1Flood11 test set. On the Sen1Flood11 Bolivia test set, our model also achieves very high mIoU (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>47.88</mn><mo>%</mo></mrow></semantics></math></inline-formula>), IoU (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>76.74</mn><mo>%</mo></mrow></semantics></math></inline-formula>), and OA (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>95.59</mn><mo>%</mo></mrow></semantics></math></inline-formula>) and shows good generalization ability.https://www.mdpi.com/2072-4292/13/11/2220deep learningSen1Floods11 datasetsmulti-source data fusionSentinel-1Sentinel-2permanent water |
| spellingShingle | Yanbing Bai Wenqi Wu Zhengxin Yang Jinze Yu Bo Zhao Xing Liu Hanfang Yang Erick Mas Shunichi Koshimura Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets Remote Sensing deep learning Sen1Floods11 datasets multi-source data fusion Sentinel-1 Sentinel-2 permanent water |
| title | Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets |
| title_full | Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets |
| title_fullStr | Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets |
| title_full_unstemmed | Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets |
| title_short | Enhancement of Detecting Permanent Water and Temporary Water in Flood Disasters by Fusing Sentinel-1 and Sentinel-2 Imagery Using Deep Learning Algorithms: Demonstration of Sen1Floods11 Benchmark Datasets |
| title_sort | enhancement of detecting permanent water and temporary water in flood disasters by fusing sentinel 1 and sentinel 2 imagery using deep learning algorithms demonstration of sen1floods11 benchmark datasets |
| topic | deep learning Sen1Floods11 datasets multi-source data fusion Sentinel-1 Sentinel-2 permanent water |
| url | https://www.mdpi.com/2072-4292/13/11/2220 |
| work_keys_str_mv | AT yanbingbai enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT wenqiwu enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT zhengxinyang enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT jinzeyu enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT bozhao enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT xingliu enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT hanfangyang enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT erickmas enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets AT shunichikoshimura enhancementofdetectingpermanentwaterandtemporarywaterinflooddisastersbyfusingsentinel1andsentinel2imageryusingdeeplearningalgorithmsdemonstrationofsen1floods11benchmarkdatasets |