Automated Extraction of Landslides Using Deep Learning and Multiple Environmental Factors

DOI

10.2991/978-94-6463-415-0_15How to use a DOI?

Keywords

Landslide; Deep learning; Taiwan; RestNet34

Abstract

As a common geological hazard, landslides pose a threat to human life and property safety. This study employs the ResNet34 method, a deep learning approach, while also considering environmental factors that influence landslides in landslide susceptibility assessment methods. The study focuses on Taiwan Island, selecting optical satellite images and 10 environmental factors to establish training and validation samples. Deep learning models can automatically extract features from sample data, a process referred to in this paper as extraction.The study utilizes optical satellite images from the entire Taiwan Island for the years 2015 and 2016, along with corresponding environmental factors such as NDVI, geology, soil, precipitation, land use, DEM, slope, aspect, planar curvature, and profile curvature. These datasets are overlaid in a band fusion manner and then input into the ResNet34 model. The experiment investigates the impact of different datasets on landslide feature extraction. Ultimately, it is found that land use data brings the most significant gain, while DEM data results in a negative impact.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yanhao Huang
PY  - 2024
DA  - 2024/05/14
TI  - Automated Extraction of Landslides Using Deep Learning and Multiple Environmental Factors
BT  - Proceedings of the 2023 9th International Conference on Advances in Energy Resources and Environment Engineering (ICAESEE 2023)
PB  - Atlantis Press
SP  - 139
EP  - 146
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-415-0_15
DO  - 10.2991/978-94-6463-415-0_15
ID  - Huang2024
ER  -