Line‐Based Levee Detection: A New Generation of Levee Detection Methods Using Digital Elevation Models

Abstract

Levee location information is an essential input for accurate flood simulation. Several methods for detecting levees from LiDAR digital elevation models have been proposed in recent years. However, these detection methods are pixel‐based, treating continuous levees as discrete points, classifying each pixel as levee or non‐levee. Thus, they produce fragmented results and struggle to balance high recall (correctly identifying levee pixels) with high precision (avoiding misclassification of non‐levee pixels). In this study, we propose a framework for a new generation of line‐based detection methods that capture the linear continuous nature of levees and simultaneously optimize precision and recall. We first develop a line detection algorithm to identify all linear features that are potentially levees from DEM data. Each line is then classified as levee or non‐levee based on statistics of pixels along the line. We find that the line detection algorithm, combined with a newly developed active learning logistic regression model, results in a line‐based method that outperforms the latest pixel‐based method across all metrics. Specifically, the F1 score increased from 0.39 to 0.59 in the United States and from 0.23 to 0.54 in New Zealand. These increases are attributable to higher precision by the line detection algorithm and higher recall by the logistic regression model. By applying the newly proposed method to 31 countries, we detected a total of 54,200 km of levees, providing a new data set for global flood risk assessment.

Using Geographic Information Systems (GIS) in Analyzing the Digital Elevation Model (DEM) for Idlib Governorate, Northwest Syria
Ahmad Al-Mousa, Badr Al-Din Al-Dakhil
Wadi Alshatti University Journal of Pure and Applied Sciences22 May 2026Peer-reviewedCC BY-NCFull text5 citedHydrology & soilsOpenAlex
Object Detection of Tree Uprooting Signatures in Gridded LiDAR Elevation Data
T. H. Doane, A. Agarwal, B. J. Yanites +1
Earth and Space Science30 Oct 2026Peer-reviewedCC BY 4.0Full textGeneral earth scienceCrossref
Impact-based flood-damage assessment in Japan under longitudinal typhoon-track perturbations and levee-height variations during Hagibis (2019)
Xiaoyang Li, Kei Yoshimura, Hironori Fudeyasu
Communications Earth & Environment13 Oct 2026Peer-reviewedCC BY 4.0Full textClimate scienceCrossref
Changepoint detection methods as a practical tool for assessment of ecological restoration
Mengyi Gong, Simone Conte, Tom Jameson +4
EcoEvoRxiv14 Oct 2026PreprintCC BY 4.0Full textEnvironmental science & pollution StatisticsEcoEvoRxiv
Micro- and nanoplastics in human biological materials: a systematic review of detection methods and methodological challenges
Aleksandra Dmitrowicz, Mariusz Sacharczuk, Dominik S. Skiba
Archives of Toxicology18 May 2026Peer-reviewedCC BY-NC-NDFull text11 citedEnvironmental science & pollutionOpenAlex

Cite

BibTeX RIS

BibTeX

@article{khanh2026linebased,
  title = {{Line‐Based Levee Detection: A New Generation of Levee Detection Methods Using Digital Elevation Models}},
  author = {Do Ngoc Khanh and Kazuma Shiraishi and Yu Yoshimura and Yukiko Hirabayashi},
  journal = {Earth and Space Science},
  year = {2026},
  month = oct,
  doi = {10.1029/2026ea005231},
  url = {https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2026EA005231},
  note = {CC BY 4.0},
}

RIS

TY  - JOUR
TI  - Line‐Based Levee Detection: A New Generation of Levee Detection Methods Using Digital Elevation Models
AU  - Do Ngoc Khanh
AU  - Kazuma Shiraishi
AU  - Yu Yoshimura
AU  - Yukiko Hirabayashi
PY  - 2026
DA  - 2026/10/30
JO  - Earth and Space Science
DO  - 10.1029/2026ea005231
UR  - https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2026EA005231
AB  - Levee location information is an essential input for accurate flood simulation. Several methods for detecting levees from LiDAR digital elevation models have been proposed in recent years. However, these detection methods are pixel‐based, treating continuous levees as discrete points, classifying each pixel as levee or non‐levee. Thus, they produce fragmented results and struggle to balance high recall (correctly identifying levee pixels) with high precision (avoiding misclassification of non‐levee pixels). In this study, we propose a framework for a new generation of line‐based detection methods that capture the linear continuous nature of levees and simultaneously optimize precision and recall. We first develop a line detection algorithm to identify all linear features that are potentially levees from DEM data. Each line is then classified as levee or non‐levee based on statistics of pixels along the line. We find that the line detection algorithm, combined with a newly developed active learning logistic regression model, results in a line‐based method that outperforms the latest pixel‐based method across all metrics. Specifically, the F1 score increased from 0.39 to 0.59 in the United States and from 0.23 to 0.54 in New Zealand. These increases are attributable to higher precision by the line detection algorithm and higher recall by the logistic regression model. By applying the newly proposed method to 31 countries, we detected a total of 54,200 km of levees, providing a new data set for global flood risk assessment.
SN  - 2333-5084
SN  - 2333-5084
ER  -

Abstract

Related

Cite