Author(s): Liu Jie; Sugihara Yuji

Linked Author(s):

Keywords: Deep learning; Heavy rain; Landslide; Risk assessment; Slope unit

Abstract: Kitakyushu City, located in the northern part of Kyushu Island, Japan, has often experienced natural disasters caused by heavy rainfall. In particular, landslides by short-duration heavy rainfall have caused enormous human and economic losses. In recent years, many studies have focused on evaluating landslide susceptibility with AI technologies. In this study, we investigated potential relationships among factors influencing landslide disasters based on a data augmentation technique for the slope unit. Moreover, through a transfer learning approach, a pixel-point-based Deep Learning Neural Network (DLNN) model was made for mapping the spatial distribution of the landslide susceptibility in the Kitakyushu area.

DOI: https://doi.org/10.64697/978-90-835589-7-4_41WC-P2023-cd

Year: 2025