Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Landslide mapping from Sentinel-2 imagery through change detection

Tommaso Monopoli, Fabio Montello, Claudio Rossi

TL;DR

This paper tackles the challenge of rapidly mapping newly triggered landslides using moderate-resolution satellite data. It reframes landslide delineation as a bi-temporal change-detection problem and introduces BBUnet, a bitemporal-bimodal network that fuses Sentinel-2 image pairs with DEM context to improve delineation. A globally diverse geodatabase of 34,920 manually validated landslide polygons is constructed from multiple inventories, enabling robust evaluation across heterogeneous ecoregions. Results show BBUnet outperforms several baselines, indicating that incorporating terrain information via DEM data can enhance landslide detection in Sentinel-2 imagery; the authors also release code and data to foster further research.

Abstract

Landslides are one of the most critical and destructive geohazards. Widespread development of human activities and settlements combined with the effects of climate change on weather are resulting in a high increase in the frequency and destructive power of landslides, making them a major threat to human life and the economy. In this paper, we explore methodologies to map newly-occurred landslides using Sentinel-2 imagery automatically. All approaches presented are framed as a bi-temporal change detection problem, requiring only a pair of Sentinel-2 images, taken respectively before and after a landslide-triggering event. Furthermore, we introduce a novel deep learning architecture for fusing Sentinel-2 bi-temporal image pairs with Digital Elevation Model (DEM) data, showcasing its promising performances w.r.t. other change detection models in the literature. As a parallel task, we address limitations in existing datasets by creating a novel geodatabase, which includes manually validated open-access landslide inventories over heterogeneous ecoregions of the world. We release both code and dataset with an open-source license.

Landslide mapping from Sentinel-2 imagery through change detection

TL;DR

This paper tackles the challenge of rapidly mapping newly triggered landslides using moderate-resolution satellite data. It reframes landslide delineation as a bi-temporal change-detection problem and introduces BBUnet, a bitemporal-bimodal network that fuses Sentinel-2 image pairs with DEM context to improve delineation. A globally diverse geodatabase of 34,920 manually validated landslide polygons is constructed from multiple inventories, enabling robust evaluation across heterogeneous ecoregions. Results show BBUnet outperforms several baselines, indicating that incorporating terrain information via DEM data can enhance landslide detection in Sentinel-2 imagery; the authors also release code and data to foster further research.

Abstract

Landslides are one of the most critical and destructive geohazards. Widespread development of human activities and settlements combined with the effects of climate change on weather are resulting in a high increase in the frequency and destructive power of landslides, making them a major threat to human life and the economy. In this paper, we explore methodologies to map newly-occurred landslides using Sentinel-2 imagery automatically. All approaches presented are framed as a bi-temporal change detection problem, requiring only a pair of Sentinel-2 images, taken respectively before and after a landslide-triggering event. Furthermore, we introduce a novel deep learning architecture for fusing Sentinel-2 bi-temporal image pairs with Digital Elevation Model (DEM) data, showcasing its promising performances w.r.t. other change detection models in the literature. As a parallel task, we address limitations in existing datasets by creating a novel geodatabase, which includes manually validated open-access landslide inventories over heterogeneous ecoregions of the world. We release both code and dataset with an open-source license.
Paper Structure (8 sections, 2 figures, 2 tables)

This paper contains 8 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Materials and methods
  4. Database creation
  5. Dataset download and processing
  6. Data augmentations
  7. Experiments and results
  8. Conclusion

Figures (2)

  • Figure 1: Sentinel-2 images before and after a landslide-triggering catastrophic event. (a)-(b)-(c): a sample from Haiti inventory. (d)-(e)-(f): a sample from Indonesia inventory. (c) and (f) show landslide polygons which are present in the respective inventories overlayed on the post image (dimmed for visibility).
  • Figure 2: BBUnet architecture, with detail on the Bitemporal-Bimodal Fusion (BBF) module.