Multi-scale differential network for landslide extraction from remote sensing images with different scenarios

Landslides are major geological hazards globally, causing significant economic losses each year. Accurate landslide detection is essential for disaster prevention, risk assessment, and timely emergency response. Current extraction methods struggle to distinguish landslides from their surroundings an...

Full description

Saved in:
Bibliographic Details
Main Authors: Bo Yu, Meijiang Zhu, Fang Chen, Ning Wang, Huichen Zhao, Lei Wang
Format: Article
Language:English
Published: Taylor & Francis Group 2024-12-01
Series:International Journal of Digital Earth
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/17538947.2024.2441920
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846116428698091520
author Bo Yu
Meijiang Zhu
Fang Chen
Ning Wang
Huichen Zhao
Lei Wang
author_facet Bo Yu
Meijiang Zhu
Fang Chen
Ning Wang
Huichen Zhao
Lei Wang
author_sort Bo Yu
collection DOAJ
description Landslides are major geological hazards globally, causing significant economic losses each year. Accurate landslide detection is essential for disaster prevention, risk assessment, and timely emergency response. Current extraction methods struggle to distinguish landslides from their surroundings and precisely define their boundaries. To address these challenges, we introduce the Multi-Scale Difference Enhancement Network (MSDENet), a framework for landslide extraction through time-based change detection. MSDENet incorporates three core components: the Difference Guided Attention Module (DGAM) for enhanced focus on landslide-specific changes, the Multi-Scale Feature Fusion Module (MSFFM) for improved boundary delineation, and the Multi-Scale Sensory Module (MSSM) to boost generalization by integrating multi-scale features. We validate MSDENet’s effectiveness on the Global Very-High-Resolution Landslide Mapping (GVLM) dataset, covering 17 diverse landslide events, and further assess its applicability on high-resolution Nepal and Wenchuan datasets. MSDENet outperforms six contemporary frameworks, achieving IoU improvements of 1.42% and 1.08% for the Kaikoura and Tbilisi datasets and demonstrating gains of 3.97% and 4.79% for the Nepal and Wenchuan datasets, confirming its effectiveness in varied conditions.
format Article
id doaj-art-e9a1591e221a46aa9466fdf2ec7a3f62
institution Kabale University
issn 1753-8947
1753-8955
language English
publishDate 2024-12-01
publisher Taylor & Francis Group
record_format Article
series International Journal of Digital Earth
spelling doaj-art-e9a1591e221a46aa9466fdf2ec7a3f622024-12-19T02:41:40ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552024-12-0117110.1080/17538947.2024.2441920Multi-scale differential network for landslide extraction from remote sensing images with different scenariosBo Yu0Meijiang Zhu1Fang Chen2Ning Wang3Huichen Zhao4Lei Wang5International Research Center of Big Data for Sustainable Development Goals, Beijing, People’s Republic of ChinaInternational Research Center of Big Data for Sustainable Development Goals, Beijing, People’s Republic of ChinaInternational Research Center of Big Data for Sustainable Development Goals, Beijing, People’s Republic of ChinaInternational Research Center of Big Data for Sustainable Development Goals, Beijing, People’s Republic of ChinaInstitute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, People’s Republic of ChinaInternational Research Center of Big Data for Sustainable Development Goals, Beijing, People’s Republic of ChinaLandslides are major geological hazards globally, causing significant economic losses each year. Accurate landslide detection is essential for disaster prevention, risk assessment, and timely emergency response. Current extraction methods struggle to distinguish landslides from their surroundings and precisely define their boundaries. To address these challenges, we introduce the Multi-Scale Difference Enhancement Network (MSDENet), a framework for landslide extraction through time-based change detection. MSDENet incorporates three core components: the Difference Guided Attention Module (DGAM) for enhanced focus on landslide-specific changes, the Multi-Scale Feature Fusion Module (MSFFM) for improved boundary delineation, and the Multi-Scale Sensory Module (MSSM) to boost generalization by integrating multi-scale features. We validate MSDENet’s effectiveness on the Global Very-High-Resolution Landslide Mapping (GVLM) dataset, covering 17 diverse landslide events, and further assess its applicability on high-resolution Nepal and Wenchuan datasets. MSDENet outperforms six contemporary frameworks, achieving IoU improvements of 1.42% and 1.08% for the Kaikoura and Tbilisi datasets and demonstrating gains of 3.97% and 4.79% for the Nepal and Wenchuan datasets, confirming its effectiveness in varied conditions.https://www.tandfonline.com/doi/10.1080/17538947.2024.2441920Remote sensingdeep learninglandslide extractionchange detection
spellingShingle Bo Yu
Meijiang Zhu
Fang Chen
Ning Wang
Huichen Zhao
Lei Wang
Multi-scale differential network for landslide extraction from remote sensing images with different scenarios
International Journal of Digital Earth
Remote sensing
deep learning
landslide extraction
change detection
title Multi-scale differential network for landslide extraction from remote sensing images with different scenarios
title_full Multi-scale differential network for landslide extraction from remote sensing images with different scenarios
title_fullStr Multi-scale differential network for landslide extraction from remote sensing images with different scenarios
title_full_unstemmed Multi-scale differential network for landslide extraction from remote sensing images with different scenarios
title_short Multi-scale differential network for landslide extraction from remote sensing images with different scenarios
title_sort multi scale differential network for landslide extraction from remote sensing images with different scenarios
topic Remote sensing
deep learning
landslide extraction
change detection
url https://www.tandfonline.com/doi/10.1080/17538947.2024.2441920
work_keys_str_mv AT boyu multiscaledifferentialnetworkforlandslideextractionfromremotesensingimageswithdifferentscenarios
AT meijiangzhu multiscaledifferentialnetworkforlandslideextractionfromremotesensingimageswithdifferentscenarios
AT fangchen multiscaledifferentialnetworkforlandslideextractionfromremotesensingimageswithdifferentscenarios
AT ningwang multiscaledifferentialnetworkforlandslideextractionfromremotesensingimageswithdifferentscenarios
AT huichenzhao multiscaledifferentialnetworkforlandslideextractionfromremotesensingimageswithdifferentscenarios
AT leiwang multiscaledifferentialnetworkforlandslideextractionfromremotesensingimageswithdifferentscenarios