A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images
Flood is one of the deadliest natural hazards worldwide, with the population affected being more than 2 billion between 1998–2017 with a lack of warning systems according to WHO. Especially, flash floods have the potential to generate fatal damages due to their rapid evolution and the limited warnin...
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
MDPI AG
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/43432e01b7624d8f9582351e7533e420 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:43432e01b7624d8f9582351e7533e420 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:43432e01b7624d8f9582351e7533e4202021-11-25T18:16:29ZA Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images10.3390/math92228462227-7390https://doaj.org/article/43432e01b7624d8f9582351e7533e4202021-11-01T00:00:00Zhttps://www.mdpi.com/2227-7390/9/22/2846https://doaj.org/toc/2227-7390Flood is one of the deadliest natural hazards worldwide, with the population affected being more than 2 billion between 1998–2017 with a lack of warning systems according to WHO. Especially, flash floods have the potential to generate fatal damages due to their rapid evolution and the limited warning and response time. An effective Early Warning Systems (EWS) could support detection and recognition of flash floods. Information about a flash flood can be mainly provided from observations of hydrology and from satellite images taken before the flash flood happens. Then, predictions from satellite images can be integrated with predictions based on sensors’ information to improve the accuracy of a forecasting system and subsequently trigger warning systems. The existing Deep Learning models such as UNET has been effectively used to segment the flash flood with high performance, but there are no ways to determine the most suitable model architecture with the proper number of layers showing the best performance in the task. In this paper, we propose a novel Deep Learning architecture, namely PSO-UNET, which combines Particle Swarm Optimization (PSO) with UNET to seek the best number of layers and the parameters of layers in the UNET based architecture; thereby improving the performance of flash flood segmentation from satellite images. Since the original UNET has a symmetrical architecture, the evolutionary computation is performed by paying attention to the contracting path and the expanding path is synchronized with the following layers in the contracting path. The UNET convolutional process is performed four times. Indeed, we consider each process as a block of the convolution having two convolutional layers in the original architecture. Training of inputs and hyper-parameters is performed by executing the PSO algorithm. In practice, the value of Dice Coefficient of our proposed model exceeds 79.75% (8.59% higher than that of the original UNET model). Experimental results on various satellite images prove the advantages and superiority of the PSO-UNET approach.Do Ngoc TuyenTran Manh TuanLe Hoang SonTran Thi NganNguyen Long GiangPham Huy ThongVu Van HieuVassilis C. GerogiannisDimitrios TzimosAndreas KanavosMDPI AGarticledeep learningParticle Swarm Optimization (PSO)UNETsatellite imagesflash flood detectionsemantic segmentationMathematicsQA1-939ENMathematics, Vol 9, Iss 2846, p 2846 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
deep learning Particle Swarm Optimization (PSO) UNET satellite images flash flood detection semantic segmentation Mathematics QA1-939 |
| spellingShingle |
deep learning Particle Swarm Optimization (PSO) UNET satellite images flash flood detection semantic segmentation Mathematics QA1-939 Do Ngoc Tuyen Tran Manh Tuan Le Hoang Son Tran Thi Ngan Nguyen Long Giang Pham Huy Thong Vu Van Hieu Vassilis C. Gerogiannis Dimitrios Tzimos Andreas Kanavos A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images |
| description |
Flood is one of the deadliest natural hazards worldwide, with the population affected being more than 2 billion between 1998–2017 with a lack of warning systems according to WHO. Especially, flash floods have the potential to generate fatal damages due to their rapid evolution and the limited warning and response time. An effective Early Warning Systems (EWS) could support detection and recognition of flash floods. Information about a flash flood can be mainly provided from observations of hydrology and from satellite images taken before the flash flood happens. Then, predictions from satellite images can be integrated with predictions based on sensors’ information to improve the accuracy of a forecasting system and subsequently trigger warning systems. The existing Deep Learning models such as UNET has been effectively used to segment the flash flood with high performance, but there are no ways to determine the most suitable model architecture with the proper number of layers showing the best performance in the task. In this paper, we propose a novel Deep Learning architecture, namely PSO-UNET, which combines Particle Swarm Optimization (PSO) with UNET to seek the best number of layers and the parameters of layers in the UNET based architecture; thereby improving the performance of flash flood segmentation from satellite images. Since the original UNET has a symmetrical architecture, the evolutionary computation is performed by paying attention to the contracting path and the expanding path is synchronized with the following layers in the contracting path. The UNET convolutional process is performed four times. Indeed, we consider each process as a block of the convolution having two convolutional layers in the original architecture. Training of inputs and hyper-parameters is performed by executing the PSO algorithm. In practice, the value of Dice Coefficient of our proposed model exceeds 79.75% (8.59% higher than that of the original UNET model). Experimental results on various satellite images prove the advantages and superiority of the PSO-UNET approach. |
| format |
article |
| author |
Do Ngoc Tuyen Tran Manh Tuan Le Hoang Son Tran Thi Ngan Nguyen Long Giang Pham Huy Thong Vu Van Hieu Vassilis C. Gerogiannis Dimitrios Tzimos Andreas Kanavos |
| author_facet |
Do Ngoc Tuyen Tran Manh Tuan Le Hoang Son Tran Thi Ngan Nguyen Long Giang Pham Huy Thong Vu Van Hieu Vassilis C. Gerogiannis Dimitrios Tzimos Andreas Kanavos |
| author_sort |
Do Ngoc Tuyen |
| title |
A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images |
| title_short |
A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images |
| title_full |
A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images |
| title_fullStr |
A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images |
| title_full_unstemmed |
A Novel Approach Combining Particle Swarm Optimization and Deep Learning for Flash Flood Detection from Satellite Images |
| title_sort |
novel approach combining particle swarm optimization and deep learning for flash flood detection from satellite images |
| publisher |
MDPI AG |
| publishDate |
2021 |
| url |
https://doaj.org/article/43432e01b7624d8f9582351e7533e420 |
| work_keys_str_mv |
AT dongoctuyen anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT tranmanhtuan anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT lehoangson anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT tranthingan anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT nguyenlonggiang anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT phamhuythong anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT vuvanhieu anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT vassiliscgerogiannis anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT dimitriostzimos anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT andreaskanavos anovelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT dongoctuyen novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT tranmanhtuan novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT lehoangson novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT tranthingan novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT nguyenlonggiang novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT phamhuythong novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT vuvanhieu novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT vassiliscgerogiannis novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT dimitriostzimos novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages AT andreaskanavos novelapproachcombiningparticleswarmoptimizationanddeeplearningforflashflooddetectionfromsatelliteimages |
| _version_ |
1718411375832727552 |