High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery

Red tides caused by <i>Margalefidinium polykrikoides</i> occur continuously along the southern coast of Korea, where there are many aquaculture cages, and therefore, prompt monitoring of bloom water is required to prevent considerable damage. Satellite-based ocean-color sensors are widel...

Full description

Bibliographic Details
Main Authors: Jisun Shin, Young-Heon Jo, Joo-Hyung Ryu, Boo-Keun Khim, Soo Mee Kim
Format: Article
Language:English
Published: MDPI AG 2021-06-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/21/13/4447
_version_ 1827600925444800512
author Jisun Shin
Young-Heon Jo
Joo-Hyung Ryu
Boo-Keun Khim
Soo Mee Kim
author_facet Jisun Shin
Young-Heon Jo
Joo-Hyung Ryu
Boo-Keun Khim
Soo Mee Kim
author_sort Jisun Shin
collection DOAJ
description Red tides caused by <i>Margalefidinium polykrikoides</i> occur continuously along the southern coast of Korea, where there are many aquaculture cages, and therefore, prompt monitoring of bloom water is required to prevent considerable damage. Satellite-based ocean-color sensors are widely used for detecting red tide blooms, but their low spatial resolution restricts coastal observations. Contrarily, terrestrial sensors with a high spatial resolution are good candidate sensors, despite the lack of spectral resolution and bands for red tide detection. In this study, we developed a U-Net deep learning model for detecting <i>M. polykrikoides</i> blooms along the southern coast of Korea from PlanetScope imagery with a high spatial resolution of 3 m. The U-Net model was trained with four different datasets that were constructed with randomly or non-randomly chosen patches consisting of different ratios of red tide and non-red tide pixels. The qualitative and quantitative assessments of the conventional red tide index (RTI) and four U-Net models suggest that the U-Net model, which was trained with a dataset of non-randomly chosen patches including non-red tide patches, outperformed RTI in terms of sensitivity, precision, and F-measure level, accounting for an increase of 19.84%, 44.84%, and 28.52%, respectively. The <i>M. polykrikoides</i> map derived from U-Net provides the most reasonable red tide patterns in all water areas. Combining high spatial resolution images and deep learning approaches represents a good solution for the monitoring of red tides over coastal regions.
first_indexed 2024-03-09T04:44:59Z
format Article
id doaj.art-751a5cfeb7d44994a7c88082675007fc
institution Directory Open Access Journal
issn 1424-8220
language English
last_indexed 2024-03-09T04:44:59Z
publishDate 2021-06-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj.art-751a5cfeb7d44994a7c88082675007fc2023-12-03T13:16:34ZengMDPI AGSensors1424-82202021-06-012113444710.3390/s21134447High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope ImageryJisun Shin0Young-Heon Jo1Joo-Hyung Ryu2Boo-Keun Khim3Soo Mee Kim4BK21 School of Earth and Environmental Systems, Pusan National University, Busan 46241, KoreaBK21 School of Earth and Environmental Systems, Pusan National University, Busan 46241, KoreaKorea Ocean Satellite Center, Korea Institute of Ocean Science and Technology (KIOST), Busan 49111, KoreaBK21 School of Earth and Environmental Systems, Pusan National University, Busan 46241, KoreaMaritime ICT R&D Center, Korea Institute of Ocean Science and Technology (KIOST), Busan 49111, KoreaRed tides caused by <i>Margalefidinium polykrikoides</i> occur continuously along the southern coast of Korea, where there are many aquaculture cages, and therefore, prompt monitoring of bloom water is required to prevent considerable damage. Satellite-based ocean-color sensors are widely used for detecting red tide blooms, but their low spatial resolution restricts coastal observations. Contrarily, terrestrial sensors with a high spatial resolution are good candidate sensors, despite the lack of spectral resolution and bands for red tide detection. In this study, we developed a U-Net deep learning model for detecting <i>M. polykrikoides</i> blooms along the southern coast of Korea from PlanetScope imagery with a high spatial resolution of 3 m. The U-Net model was trained with four different datasets that were constructed with randomly or non-randomly chosen patches consisting of different ratios of red tide and non-red tide pixels. The qualitative and quantitative assessments of the conventional red tide index (RTI) and four U-Net models suggest that the U-Net model, which was trained with a dataset of non-randomly chosen patches including non-red tide patches, outperformed RTI in terms of sensitivity, precision, and F-measure level, accounting for an increase of 19.84%, 44.84%, and 28.52%, respectively. The <i>M. polykrikoides</i> map derived from U-Net provides the most reasonable red tide patterns in all water areas. Combining high spatial resolution images and deep learning approaches represents a good solution for the monitoring of red tides over coastal regions.https://www.mdpi.com/1424-8220/21/13/4447<i>Margalefidinium polykrikoides</i>PlanetScopesouthern coast of Koreaconvolutional neural networkU-Net
spellingShingle Jisun Shin
Young-Heon Jo
Joo-Hyung Ryu
Boo-Keun Khim
Soo Mee Kim
High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery
Sensors
<i>Margalefidinium polykrikoides</i>
PlanetScope
southern coast of Korea
convolutional neural network
U-Net
title High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery
title_full High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery
title_fullStr High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery
title_full_unstemmed High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery
title_short High Spatial-Resolution Red Tide Detection in the Southern Coast of Korea Using U-Net from PlanetScope Imagery
title_sort high spatial resolution red tide detection in the southern coast of korea using u net from planetscope imagery
topic <i>Margalefidinium polykrikoides</i>
PlanetScope
southern coast of Korea
convolutional neural network
U-Net
url https://www.mdpi.com/1424-8220/21/13/4447
work_keys_str_mv AT jisunshin highspatialresolutionredtidedetectioninthesoutherncoastofkoreausingunetfromplanetscopeimagery
AT youngheonjo highspatialresolutionredtidedetectioninthesoutherncoastofkoreausingunetfromplanetscopeimagery
AT joohyungryu highspatialresolutionredtidedetectioninthesoutherncoastofkoreausingunetfromplanetscopeimagery
AT bookeunkhim highspatialresolutionredtidedetectioninthesoutherncoastofkoreausingunetfromplanetscopeimagery
AT soomeekim highspatialresolutionredtidedetectioninthesoutherncoastofkoreausingunetfromplanetscopeimagery