ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture

Abstract Landslides, as a prevalent geological hazard, pose a severe threat to both the environment and human society. The rapid and accurate identification of landslide-prone areas is crucial for disaster response, risk assessment, and urban planning. This paper proposes a novel deep learning-based...

Full description

Saved in:
Bibliographic Details
Main Authors: Xuqing Ren, Xu Wu, Donghao Zhai, Xiangpeng Wang, Ningbo He, Mehreen Tarif
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-98230-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849699302545293312
author Xuqing Ren
Xu Wu
Donghao Zhai
Xiangpeng Wang
Ningbo He
Mehreen Tarif
author_facet Xuqing Ren
Xu Wu
Donghao Zhai
Xiangpeng Wang
Ningbo He
Mehreen Tarif
author_sort Xuqing Ren
collection DOAJ
description Abstract Landslides, as a prevalent geological hazard, pose a severe threat to both the environment and human society. The rapid and accurate identification of landslide-prone areas is crucial for disaster response, risk assessment, and urban planning. This paper proposes a novel deep learning-based landslide detection method ResM-FusionNet, which leverages ResNet-50 as the backbone for feature extraction and integrates a multi-layer perceptron as the decoder to enhance segmentation accuracy. To address the challenges of complex terrain and boundary detail detection in landslide-prone regions, we introduce a novel loss function, RLoss, which integrates masking mechanisms, semantic weighting, and a Top-K pixel loss averaging strategy. Experimental results on remote sensing datasets demonstrate that ResM-FusionNet significantly outperforms existing models. Specifically, ResM-FusionNet achieves 94.33% accuracy, 85.73% F1-score, and a Kappa coefficient of 70.12%, surpassing other models (e.g., SegFormer, DeepLabv3, and UNet) by 4.4%, 7.7%, and 17.6% in accuracy, respectively. Moreover, ResM-FusionNet excels in boundary detection, achieving an IoU of 0.7545, precision of 85.61%, and recall of 83.92%. These findings indicate that the proposed method not only provides robust and accurate landslide detection but also enhances segmentation performance in complex terrains.
format Article
id doaj-art-2a0bb2df16d0474cbc760f809acb054c
institution DOAJ
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-2a0bb2df16d0474cbc760f809acb054c2025-08-20T03:18:38ZengNature PortfolioScientific Reports2045-23222025-04-0115111910.1038/s41598-025-98230-6ResM-FusionNet for efficient landslide detection algorithm with a hybrid architectureXuqing Ren0Xu Wu1Donghao Zhai2Xiangpeng Wang3Ningbo He4Mehreen Tarif5College of Computers Science and Cyber Security, Chengdu University of TechnologyCollege of Computers Science and Cyber Security, Chengdu University of TechnologyCollege of Computers Science and Cyber Security, Chengdu University of TechnologyCollege of Geophysics, Chengdu University of TechnologyChina ANNENG Group Third Engineering BureauCollege of Computers Science and Cyber Security, Chengdu University of TechnologyAbstract Landslides, as a prevalent geological hazard, pose a severe threat to both the environment and human society. The rapid and accurate identification of landslide-prone areas is crucial for disaster response, risk assessment, and urban planning. This paper proposes a novel deep learning-based landslide detection method ResM-FusionNet, which leverages ResNet-50 as the backbone for feature extraction and integrates a multi-layer perceptron as the decoder to enhance segmentation accuracy. To address the challenges of complex terrain and boundary detail detection in landslide-prone regions, we introduce a novel loss function, RLoss, which integrates masking mechanisms, semantic weighting, and a Top-K pixel loss averaging strategy. Experimental results on remote sensing datasets demonstrate that ResM-FusionNet significantly outperforms existing models. Specifically, ResM-FusionNet achieves 94.33% accuracy, 85.73% F1-score, and a Kappa coefficient of 70.12%, surpassing other models (e.g., SegFormer, DeepLabv3, and UNet) by 4.4%, 7.7%, and 17.6% in accuracy, respectively. Moreover, ResM-FusionNet excels in boundary detection, achieving an IoU of 0.7545, precision of 85.61%, and recall of 83.92%. These findings indicate that the proposed method not only provides robust and accurate landslide detection but also enhances segmentation performance in complex terrains.https://doi.org/10.1038/s41598-025-98230-6Landslide detectionDeep learningResNet-50Multilayer perceptronLoss function RLoss
spellingShingle Xuqing Ren
Xu Wu
Donghao Zhai
Xiangpeng Wang
Ningbo He
Mehreen Tarif
ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture
Scientific Reports
Landslide detection
Deep learning
ResNet-50
Multilayer perceptron
Loss function RLoss
title ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture
title_full ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture
title_fullStr ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture
title_full_unstemmed ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture
title_short ResM-FusionNet for efficient landslide detection algorithm with a hybrid architecture
title_sort resm fusionnet for efficient landslide detection algorithm with a hybrid architecture
topic Landslide detection
Deep learning
ResNet-50
Multilayer perceptron
Loss function RLoss
url https://doi.org/10.1038/s41598-025-98230-6
work_keys_str_mv AT xuqingren resmfusionnetforefficientlandslidedetectionalgorithmwithahybridarchitecture
AT xuwu resmfusionnetforefficientlandslidedetectionalgorithmwithahybridarchitecture
AT donghaozhai resmfusionnetforefficientlandslidedetectionalgorithmwithahybridarchitecture
AT xiangpengwang resmfusionnetforefficientlandslidedetectionalgorithmwithahybridarchitecture
AT ningbohe resmfusionnetforefficientlandslidedetectionalgorithmwithahybridarchitecture
AT mehreentarif resmfusionnetforefficientlandslidedetectionalgorithmwithahybridarchitecture