Landslide Susceptibility Assessment Based on a Deep Learning-Derived Landslide Inventory in Moxi Town, Sichuan, China
Remote Sensing · 10.03.2026 · 18 (6) · 849
DOI: 10.3390/rs18060849 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 10.03.2026. Verfasst von Yitong Yao und 9 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Sichuan Science and Technology Program, Science and Technology Open Foundation of Sichuan Society of Surveying and Mapping Geoinformation.
Abstract (Originalsprache)
Landslides are characterized by strong suddenness and a wide range of damage; accurate prediction of their susceptibility is an important prerequisite for regional risk prevention and control. To address the difficulties in acquiring landslide inventories in complex terrain areas and the insufficient interpretability of existing prediction models, this study proposes a landslide susceptibility assessment (LSA) framework that integrates automated sample detection and interpretability analysis. The proposed framework is applied to Moxi Town, a typical alpine valley area in Sichuan Province, China. A Mask R-CNN instance segmentation model was introduced to achieve automated detection of landslide samples, resulting in a high-quality inventory containing 923 landslides. Based on the spatial relationships between the landslide inventory and influencing factors, a convolutional neural network (CNN) landslide susceptibility assessment model incorporating Shapley Additive exPlanations (SHAP) interpretability analysis was constructed. The CNN model was further compared with random forest (RF) and extreme gradient boosting (XGBoost) machine learning models. The results show that the AUC value of the CNN model has increased by 4.3% and 3.2% compared with the RF and XGBoost models, respectively, and it significantly reduces the pretzel effect of landslide susceptibility mapping (LSM). The results validate the reliability of the proposed framework, which can provide technical support for landslide disaster prevention and monitoring.
Förderung
- National Natural Science Foundation of China
- Sichuan Science and Technology Program
- Science and Technology Open Foundation of Sichuan Society of Surveying and Mapping Geoinformation
- Central Government Guided Local Science and Technology Development Fund Projects
- Supply–Demand Matching for Employment and Talent Cultivation Project of the Ministry of Education of China
- Technology Innovation Center for Emergency Surveying and Mapping
Zitieren
APA
Yao, Y., Du, Y., Zhang, W., Liu, X., Cai, J., Feng, H., Xiang, H., Hu, R., Yang, Y., & Fu, T. (2026). Landslide Susceptibility Assessment Based on a Deep Learning-Derived Landslide Inventory in Moxi Town, Sichuan, China. Remote Sensing, 18(6), 849. https://doi.org/10.3390/rs18060849
Vancouver
Yao Y, Du Y, Zhang W, Liu X, Cai J, Feng H, et al. Landslide Susceptibility Assessment Based on a Deep Learning-Derived Landslide Inventory in Moxi Town, Sichuan, China. Remote Sensing. 2026;18(6):849. doi:10.3390/rs18060849
BibTeX
@article{Yao2026,
author = {Yao, Yitong and Du, Yixiang and Zhang, Wenjun and Liu, Xianwen and Cai, Jialun and Feng, Hui and Xiang, Hongyao and Hu, Rong and Yang, Yuhao and Fu, Tongben},
title = {Landslide Susceptibility Assessment Based on a Deep Learning-Derived Landslide Inventory in Moxi Town, Sichuan, China},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {6},
pages = {849},
doi = {10.3390/rs18060849}
}
EUVIMED