Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction
Fog generally forms at dawn and dusk, which exerts serious impacts on public traffic and human health. Terrain strongly affects fog formation, which provides a useful clue for fog detection from satellite observation. With the aid of the advanced Himawari-8 imager data (H8/AHI), this study develops...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/17/4328 |
| _version_ | 1797493283982671872 |
|---|---|
| author | Yinze Ran Huiyun Ma Zengwei Liu Xiaojing Wu Yanan Li Huihui Feng |
| author_facet | Yinze Ran Huiyun Ma Zengwei Liu Xiaojing Wu Yanan Li Huihui Feng |
| author_sort | Yinze Ran |
| collection | DOAJ |
| description | Fog generally forms at dawn and dusk, which exerts serious impacts on public traffic and human health. Terrain strongly affects fog formation, which provides a useful clue for fog detection from satellite observation. With the aid of the advanced Himawari-8 imager data (H8/AHI), this study develops a deep learning algorithm for fog detection at dawn and dusk under terrain-restriction and enhanced channel domain attention mechanism (DDF-Net). The DDF-Net is based on the traditional U-Net model, with the digital elevation model (DEM) data acting as the auxiliary information to separate fog from the low stratus. Furthermore, the squeeze-and-excitation networks (SE-Net) is integrated to optimize the information extraction for eliminating the influence of solar zenith angles (SZA) on the spectral characteristics over a large region. Results show acceptable accuracy of the DDF-Net. The overall probability of detection (POD) is 84.0% at dawn and 83.7% at dusk. In addition, the terrain-restriction strategy improves the results at the edges of foggy regions and reduces the false alarm rate (FAR) for low stratus. The accuracy is expected to be improved when training at a season or month scale, rather than at a longer temporal scale. Results of our study help to improve the accuracy of fog detection, which could further support the relevant traffic planning or healthy travel. |
| first_indexed | 2024-03-10T01:17:48Z |
| format | Article |
| id | doaj.art-1ce7600a25df432c9f271c4eb7be177e |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T01:17:48Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-1ce7600a25df432c9f271c4eb7be177e2023-11-23T14:04:46ZengMDPI AGRemote Sensing2072-42922022-09-011417432810.3390/rs14174328Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-RestrictionYinze Ran0Huiyun Ma1Zengwei Liu2Xiaojing Wu3Yanan Li4Huihui Feng5School of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaNational Satellite Meteorological Center, China Meteorological Administration, Beijing 100081, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha 410083, ChinaFog generally forms at dawn and dusk, which exerts serious impacts on public traffic and human health. Terrain strongly affects fog formation, which provides a useful clue for fog detection from satellite observation. With the aid of the advanced Himawari-8 imager data (H8/AHI), this study develops a deep learning algorithm for fog detection at dawn and dusk under terrain-restriction and enhanced channel domain attention mechanism (DDF-Net). The DDF-Net is based on the traditional U-Net model, with the digital elevation model (DEM) data acting as the auxiliary information to separate fog from the low stratus. Furthermore, the squeeze-and-excitation networks (SE-Net) is integrated to optimize the information extraction for eliminating the influence of solar zenith angles (SZA) on the spectral characteristics over a large region. Results show acceptable accuracy of the DDF-Net. The overall probability of detection (POD) is 84.0% at dawn and 83.7% at dusk. In addition, the terrain-restriction strategy improves the results at the edges of foggy regions and reduces the false alarm rate (FAR) for low stratus. The accuracy is expected to be improved when training at a season or month scale, rather than at a longer temporal scale. Results of our study help to improve the accuracy of fog detection, which could further support the relevant traffic planning or healthy travel.https://www.mdpi.com/2072-4292/14/17/4328H8/AHIfog detection at dawn and duskDEMU-NetSE-Net |
| spellingShingle | Yinze Ran Huiyun Ma Zengwei Liu Xiaojing Wu Yanan Li Huihui Feng Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction Remote Sensing H8/AHI fog detection at dawn and dusk DEM U-Net SE-Net |
| title | Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction |
| title_full | Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction |
| title_fullStr | Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction |
| title_full_unstemmed | Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction |
| title_short | Satellite Fog Detection at Dawn and Dusk Based on the Deep Learning Algorithm under Terrain-Restriction |
| title_sort | satellite fog detection at dawn and dusk based on the deep learning algorithm under terrain restriction |
| topic | H8/AHI fog detection at dawn and dusk DEM U-Net SE-Net |
| url | https://www.mdpi.com/2072-4292/14/17/4328 |
| work_keys_str_mv | AT yinzeran satellitefogdetectionatdawnandduskbasedonthedeeplearningalgorithmunderterrainrestriction AT huiyunma satellitefogdetectionatdawnandduskbasedonthedeeplearningalgorithmunderterrainrestriction AT zengweiliu satellitefogdetectionatdawnandduskbasedonthedeeplearningalgorithmunderterrainrestriction AT xiaojingwu satellitefogdetectionatdawnandduskbasedonthedeeplearningalgorithmunderterrainrestriction AT yananli satellitefogdetectionatdawnandduskbasedonthedeeplearningalgorithmunderterrainrestriction AT huihuifeng satellitefogdetectionatdawnandduskbasedonthedeeplearningalgorithmunderterrainrestriction |