서브메뉴
검색
Characterization of Natural Hazards in the Great Lakes: Remote Monitoring and Deep Learning
Characterization of Natural Hazards in the Great Lakes: Remote Monitoring and Deep Learning
Detailed Information
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105158
- ISBN
- 9798293872459
- DDC
- 333.91
- 저자명
- Wang, Wei.
- 서명/저자
- Characterization of Natural Hazards in the Great Lakes: Remote Monitoring and Deep Learning
- 발행사항
- [Sl] : The University of Wisconsin - Madison, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 195 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-03, Section: B.
- 주기사항
- Advisor: Block, Paul.
- 학위논문주기
- Thesis (Ph.D.)--The University of Wisconsin - Madison, 2025.
- 초록/해제
- 요약Natural hazards are environmental events that pose significant risks to societies and human environments. In the Great Lakes region, natural hazards such as flooding, dangerous currents, and rapid shoreline changes are common concerns. These hazards have led to numerous incidents, resulting in fatalities, socio-economic losses, and ecological damage. Given the severe consequences of natural hazards, the Great Lakes region faces two primary challenges: effectively detecting these hazards with high spatial resolution while minimizing labor and time costs; and comprehensively characterizing their occurrence and features to support informed management and mitigation efforts. I hypothesize that combining remote sensing and deep learning techniques can enhance the detection and characterization of natural hazards in the Great Lakes region. To examine this hypothesis, the overall objective of this study is characterizing natural hazards using remote monitoring and deep learning methods. Specifically, the research investigates flood impacts on stream habitat quality, and flash rip currents and rapid shoreline changes in coastal areas. First, to assess flood impacts on stream habitat quality, a UAV-based toolkit was developed to characterize stream habitat quality conditions using multi-metric indices (MMIs). Applied before and after the August 2018 flood, this approach revealed patterns of loss and resilience in riparian vegetation, bank stability, and in-stream cover. Second, for flash rip currents, webcam imagery was analyzed with a refined Cascade R-CNN model, enabling reliable detection and classification of flash rips into three driving factors: water-level fluctuations, normal waves, and oblique waves. Their spatial, temporal, and kinematic features were then characterized to quantify differences among driving factors. Third, to detect and characterize rapid shoreline changes, a DeepLab-based segmentation framework was applied to aerial images for shoreline extraction and coupled with an improved Digital Shoreline Analysis System to compute change rates, identify hotspots, and distinguish true morphological change from shoreline retreats driven by water-level fluctuations. Overall, this research contributes to improving detection methodologies and enriching the characterization of natural hazards in the Great Lakes region. The findings aim to reduce hazard-related risks and provide valuable insights into effective management and mitigation efforts.
- 일반주제명
- Geomorphology
- 일반주제명
- Computer science
- 키워드
- Characterization
- 키워드
- Deep learning
- 키워드
- Detection
- 키워드
- Great Lakes
- 키워드
- Natural hazards
- 키워드
- Remote sensing
- 기타저자
- The University of Wisconsin - Madison Civil & Environmental Engr
- 기본자료저록
- Dissertations Abstracts International. 87-03B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017359682
■00520260202105158
■006m o d
■007cr#unu||||||||
■020 ▼a9798293872459
■035 ▼a(MiAaPQ)AAI32243584
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a333.91
■1001 ▼aWang, Wei.
■24510▼aCharacterization of Natural Hazards in the Great Lakes: Remote Monitoring and Deep Learning
■260 ▼a[Sl]▼bThe University of Wisconsin - Madison▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a195 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-03, Section: B.
■500 ▼aAdvisor: Block, Paul.
■5021 ▼aThesis (Ph.D.)--The University of Wisconsin - Madison, 2025.
■520 ▼aNatural hazards are environmental events that pose significant risks to societies and human environments. In the Great Lakes region, natural hazards such as flooding, dangerous currents, and rapid shoreline changes are common concerns. These hazards have led to numerous incidents, resulting in fatalities, socio-economic losses, and ecological damage. Given the severe consequences of natural hazards, the Great Lakes region faces two primary challenges: effectively detecting these hazards with high spatial resolution while minimizing labor and time costs; and comprehensively characterizing their occurrence and features to support informed management and mitigation efforts. I hypothesize that combining remote sensing and deep learning techniques can enhance the detection and characterization of natural hazards in the Great Lakes region. To examine this hypothesis, the overall objective of this study is characterizing natural hazards using remote monitoring and deep learning methods. Specifically, the research investigates flood impacts on stream habitat quality, and flash rip currents and rapid shoreline changes in coastal areas. First, to assess flood impacts on stream habitat quality, a UAV-based toolkit was developed to characterize stream habitat quality conditions using multi-metric indices (MMIs). Applied before and after the August 2018 flood, this approach revealed patterns of loss and resilience in riparian vegetation, bank stability, and in-stream cover. Second, for flash rip currents, webcam imagery was analyzed with a refined Cascade R-CNN model, enabling reliable detection and classification of flash rips into three driving factors: water-level fluctuations, normal waves, and oblique waves. Their spatial, temporal, and kinematic features were then characterized to quantify differences among driving factors. Third, to detect and characterize rapid shoreline changes, a DeepLab-based segmentation framework was applied to aerial images for shoreline extraction and coupled with an improved Digital Shoreline Analysis System to compute change rates, identify hotspots, and distinguish true morphological change from shoreline retreats driven by water-level fluctuations. Overall, this research contributes to improving detection methodologies and enriching the characterization of natural hazards in the Great Lakes region. The findings aim to reduce hazard-related risks and provide valuable insights into effective management and mitigation efforts.
■590 ▼aSchool code: 0262.
■650 4▼aWater resources management
■650 4▼aGeomorphology
■650 4▼aComputer science
■650 4▼aEnvironmental engineering
■653 ▼aCharacterization
■653 ▼aDeep learning
■653 ▼aDetection
■653 ▼aGreat Lakes
■653 ▼aNatural hazards
■653 ▼aRemote sensing
■690 ▼a0595
■690 ▼a0484
■690 ▼a0984
■690 ▼a0543
■690 ▼a0775
■71020▼aThe University of Wisconsin - Madison▼bCivil & Environmental Engr.
■7730 ▼tDissertations Abstracts International▼g87-03B.
■790 ▼a0262
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17359682▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
Preview
Export
ChatGPT Discussion
AI Recommended Related Books
Подробнее информация.
- Бронирование
- не существует
- моя папка
- Первый запрос зрения
- Non-Book Loan Application
- Nighttime Book Loan Application
Available after logging in.


