A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning

Increasing global plastic usage has raised critical concerns regarding marine pollution. This study addresses the pressing issue of floating marine macro-litter (FMML) by developing a novel monitoring system using a multi-spectral sensor and drones along the southern coast of South Korea. Subsequent...

Full description

Saved in:
Bibliographic Details
Main Authors: Youchul Jeong, Jisun Shin, Jong-Seok Lee, Ji-Yeon Baek, Daniel Schläpfer, Sin-Young Kim, Jin-Yong Jeong, Young-Heon Jo
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/16/23/4347
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846123935789219840
author Youchul Jeong
Jisun Shin
Jong-Seok Lee
Ji-Yeon Baek
Daniel Schläpfer
Sin-Young Kim
Jin-Yong Jeong
Young-Heon Jo
author_facet Youchul Jeong
Jisun Shin
Jong-Seok Lee
Ji-Yeon Baek
Daniel Schläpfer
Sin-Young Kim
Jin-Yong Jeong
Young-Heon Jo
author_sort Youchul Jeong
collection DOAJ
description Increasing global plastic usage has raised critical concerns regarding marine pollution. This study addresses the pressing issue of floating marine macro-litter (FMML) by developing a novel monitoring system using a multi-spectral sensor and drones along the southern coast of South Korea. Subsequently, a convolutional neural network (CNN) model was utilized to classify four distinct marine litter materials: film, fiber, fragment, and foam. Automatic atmospheric correction with the drone data atmospheric correction (DROACOR) method, which is specifically designed for currently available drone-based sensors, ensured consistent reflectance across altitudes in the FMML dataset. The CNN models exhibited promising performance, with precision, recall, and F1 score values of 0.9, 0.88, and 0.89, respectively. Furthermore, gradient-weighted class activation mapping (Grad-CAM), an object recognition technique, allowed us to interpret the classification performance. Overall, this study will shed light on successful FMML identification using multi-spectral observations for broader applications in diverse marine environments.
format Article
id doaj-art-b6906fab00124a069f86c0733c3d7d30
institution Kabale University
issn 2072-4292
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-b6906fab00124a069f86c0733c3d7d302024-12-13T16:30:30ZengMDPI AGRemote Sensing2072-42922024-11-011623434710.3390/rs16234347A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep LearningYouchul Jeong0Jisun Shin1Jong-Seok Lee2Ji-Yeon Baek3Daniel Schläpfer4Sin-Young Kim5Jin-Yong Jeong6Young-Heon Jo7Marine Research Institute, Pusan National University, Busan 46241, Republic of KoreaMarine Research Institute, Pusan National University, Busan 46241, Republic of KoreaBK21 School of Earth and Environmental Systems, Pusan National University, Busan 46241, Republic of KoreaBK21 School of Earth and Environmental Systems, Pusan National University, Busan 46241, Republic of KoreaReSe Applications LLC, 9500 Wil, SwitzerlandBK21 School of Earth and Environmental Systems, Pusan National University, Busan 46241, Republic of KoreaMarine Disaster Research Department, Korea Institute of Ocean Science and Technology, Busan 49111, Republic of KoreaMarine Research Institute, Pusan National University, Busan 46241, Republic of KoreaIncreasing global plastic usage has raised critical concerns regarding marine pollution. This study addresses the pressing issue of floating marine macro-litter (FMML) by developing a novel monitoring system using a multi-spectral sensor and drones along the southern coast of South Korea. Subsequently, a convolutional neural network (CNN) model was utilized to classify four distinct marine litter materials: film, fiber, fragment, and foam. Automatic atmospheric correction with the drone data atmospheric correction (DROACOR) method, which is specifically designed for currently available drone-based sensors, ensured consistent reflectance across altitudes in the FMML dataset. The CNN models exhibited promising performance, with precision, recall, and F1 score values of 0.9, 0.88, and 0.89, respectively. Furthermore, gradient-weighted class activation mapping (Grad-CAM), an object recognition technique, allowed us to interpret the classification performance. Overall, this study will shed light on successful FMML identification using multi-spectral observations for broader applications in diverse marine environments.https://www.mdpi.com/2072-4292/16/23/4347floating marine macro-litterunmanned aerial vehiclemulti-spectral sensoratmospheric correctionreflectance retrievalconvolutional neural network
spellingShingle Youchul Jeong
Jisun Shin
Jong-Seok Lee
Ji-Yeon Baek
Daniel Schläpfer
Sin-Young Kim
Jin-Yong Jeong
Young-Heon Jo
A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning
Remote Sensing
floating marine macro-litter
unmanned aerial vehicle
multi-spectral sensor
atmospheric correction
reflectance retrieval
convolutional neural network
title A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning
title_full A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning
title_fullStr A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning
title_full_unstemmed A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning
title_short A Study on the Monitoring of Floating Marine Macro-Litter Using a Multi-Spectral Sensor and Classification Based on Deep Learning
title_sort study on the monitoring of floating marine macro litter using a multi spectral sensor and classification based on deep learning
topic floating marine macro-litter
unmanned aerial vehicle
multi-spectral sensor
atmospheric correction
reflectance retrieval
convolutional neural network
url https://www.mdpi.com/2072-4292/16/23/4347
work_keys_str_mv AT youchuljeong astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jisunshin astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jongseoklee astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jiyeonbaek astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT danielschlapfer astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT sinyoungkim astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jinyongjeong astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT youngheonjo astudyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT youchuljeong studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jisunshin studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jongseoklee studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jiyeonbaek studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT danielschlapfer studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT sinyoungkim studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT jinyongjeong studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning
AT youngheonjo studyonthemonitoringoffloatingmarinemacrolitterusingamultispectralsensorandclassificationbasedondeeplearning