東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Large Scale Identification and Monit...

Sasaki, Kenichi,

Linked to FindBook

Google Book

Amazon

博客來

Large Scale Identification and Monitoring of Marine Debris Using Very High Resolution Satellite Images /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Large Scale Identification and Monitoring of Marine Debris Using Very High Resolution Satellite Images // Kenichi Sasaki.
Author:	Sasaki, Kenichi,
Description:	1 electronic resource (151 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
Contained By:	Dissertations Abstracts International85-11B.
Subject:	Remote sensing. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30995206
ISBN:	9798382719290

Large Scale Identification and Monitoring of Marine Debris Using Very High Resolution Satellite Images /
Sasaki, Kenichi,

Large Scale Identification and Monitoring of Marine Debris Using Very High Resolution Satellite Images /Kenichi Sasaki. - 1 electronic resource (151 pages)

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

Marine debris has been a serious problem for the marine ecosystem and aquatic industries. While many studies have been conducted to characterize marine debris, most of them have focused on heavily contaminated regions due to the difficulty of acquiring accurate ground truth data. They did not consider more practical situations to apply their approach to real scenes. To address this issue, this study proposes a methodology to quantitatively analyze and evaluate the debris accumulation in coastal regions using remote sensing data under practical scenarios.We acquired in-situ measurements of debris amounts in association with a local government, which is the crucial data used as a ground truth of the debris accumulation. We also amassed a large number of remote sensing data roughly coincident with the in-situ measurements including satellite images from various platforms and drone pictures.We developed a novel approach to estimate debris concentration in coastal areas from very high resolution satellite images such as WorldView images provided by MAXAR corp. using machine learning techniques. Then, we extended our analysis procedure to satellite images from different platforms using domain adaptation techniques. This extension makes it possible to build a multi-modal analysis model and increase the sampling frequency to visualize the temporal evolution of the debris deposition.We also conducted a time series analysis of satellite images to investigate the temporal evolution of debris deposits. We showed that debris migration is heavily influenced by seasonal weather changes, meteorological events and geographical factors. The resulting statistical algorithm that connects the satellite image features with the ground-truth measurements can then be used to compute debris amounts in regions with no ground truth observations. Our approach has the potential to provide a more comprehensive understanding of the dynamics of debris accumulation in coastal regions, thereby facilitating effective mitigation strategies to reduce economic impact and environmental pollution.

English

ISBN: 9798382719290Subjects--Topical Terms:

535394
Remote sensing.
Subjects--Index Terms:

Machine learning

Large Scale Identification and Monitoring of Marine Debris Using Very High Resolution Satellite Images /
LDR:03595nmm a22004333i 4500 001 2429145
005 20250430093534.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798382719290
035 $a (MiAaPQD)AAI30995206
035 $a AAI30995206
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Sasaki, Kenichi, $e author. $0 (orcid)0000-0002-6492-2245 $3 3816818
245 1 0 $a Large Scale Identification and Monitoring of Marine Debris Using Very High Resolution Satellite Images / $c Kenichi Sasaki.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (151 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
500 $a Advisors: Emery, William Committee members: Matsuo, Tomoko; Morton, Jade; Pacifici, Fabio; Moulton, Mellisa.
502 $b Ph.D. $c University of Colorado at Boulder $d 2024.
520 $a Marine debris has been a serious problem for the marine ecosystem and aquatic industries. While many studies have been conducted to characterize marine debris, most of them have focused on heavily contaminated regions due to the difficulty of acquiring accurate ground truth data. They did not consider more practical situations to apply their approach to real scenes. To address this issue, this study proposes a methodology to quantitatively analyze and evaluate the debris accumulation in coastal regions using remote sensing data under practical scenarios.We acquired in-situ measurements of debris amounts in association with a local government, which is the crucial data used as a ground truth of the debris accumulation. We also amassed a large number of remote sensing data roughly coincident with the in-situ measurements including satellite images from various platforms and drone pictures.We developed a novel approach to estimate debris concentration in coastal areas from very high resolution satellite images such as WorldView images provided by MAXAR corp. using machine learning techniques. Then, we extended our analysis procedure to satellite images from different platforms using domain adaptation techniques. This extension makes it possible to build a multi-modal analysis model and increase the sampling frequency to visualize the temporal evolution of the debris deposition.We also conducted a time series analysis of satellite images to investigate the temporal evolution of debris deposits. We showed that debris migration is heavily influenced by seasonal weather changes, meteorological events and geographical factors. The resulting statistical algorithm that connects the satellite image features with the ground-truth measurements can then be used to compute debris amounts in regions with no ground truth observations. Our approach has the potential to provide a more comprehensive understanding of the dynamics of debris accumulation in coastal regions, thereby facilitating effective mitigation strategies to reduce economic impact and environmental pollution.
546 $a English
590 $a School code: 0051
650 4 $a Remote sensing. $3 535394
650 4 $a Marine geology. $3 3173821
650 4 $a Ocean engineering. $3 660731
653 $a Machine learning
653 $a Marine debris
653 $a Multispectral images
653 $a Oceanography
653 $a Satellite imagery analysis
690 $a 0799
690 $a 0547
690 $a 0556
710 2 $a University of Colorado at Boulder. $b Aerospace Engineering. $e degree granting institution. $3 3814664
720 1 $a Emery, William $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 85-11B.
790 $a 0051
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30995206