Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Landslide detection from bitemporal satellite imagery using attention-based deep neural networks

Amankwah, Solomon Obiri Yeboah ; Wang, Guojie ; Gnyawali, Kaushal ; Hagan, Daniel Fiifi Tawiah ; Sarfo, Isaac ; Zhen, Dong ; Nooni, Isaac Kwesi ; Ullah, Waheed and Duan, Zheng LU (2022) In Landslides 19(10). p.2459-2471
Abstract

Torrential rainfall predisposes hills to catastrophic landslides resulting in serious damage to life and property. Landslide inventory maps are therefore essential for rapid response and developing disaster mitigation strategies. Manual mapping techniques are laborious and time-consuming, and thus not ideal in rapid response situations. Automated landslide mapping from optical satellite imagery using deep neural networks (DNNs) is becoming popular. However, distinguishing landslides from other changed objects in optical imagery using backbone DNNs alone is difficult. Attention modules have been introduced recently into the architecture of DNNs to address this problem by improving the discriminative ability of DNNs and suppressing noisy... (More)

Torrential rainfall predisposes hills to catastrophic landslides resulting in serious damage to life and property. Landslide inventory maps are therefore essential for rapid response and developing disaster mitigation strategies. Manual mapping techniques are laborious and time-consuming, and thus not ideal in rapid response situations. Automated landslide mapping from optical satellite imagery using deep neural networks (DNNs) is becoming popular. However, distinguishing landslides from other changed objects in optical imagery using backbone DNNs alone is difficult. Attention modules have been introduced recently into the architecture of DNNs to address this problem by improving the discriminative ability of DNNs and suppressing noisy backgrounds. This study compares two state-of-the-art attention-boosted deep Siamese neural networks in mapping rainfall-induced landslides in the mountainous Himalayan region of Nepal using Planetscope (PS) satellite imagery. Our findings confirm that attention networks improve the performance of DNNs as they can extract more discriminative features. The Siamese Nested U-Net (SNUNet) produced the best and most coherent landslide inventory map among the methods in the test area, achieving an F1-score of 0.73, which is comparable to other similar studies. Our findings demonstrate a prospect for application of the attention-based DNNs in rapid landslide mapping and disaster mitigation not only for rainfall-triggered landslides but also for earthquake-triggered landslides.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Attention module, Change detection, Deep neural network (DNN), Landslide mapping
in
Landslides
volume
19
issue
10
pages
13 pages
publisher
Springer
external identifiers
  • scopus:85132312895
ISSN
1612-510X
DOI
10.1007/s10346-022-01915-6
language
English
LU publication?
yes
id
40d2e2c4-bd14-402f-b259-55961c7e0a7f
date added to LUP
2022-09-29 13:05:19
date last changed
2023-05-16 13:07:46
@article{40d2e2c4-bd14-402f-b259-55961c7e0a7f,
  abstract     = {{<p>Torrential rainfall predisposes hills to catastrophic landslides resulting in serious damage to life and property. Landslide inventory maps are therefore essential for rapid response and developing disaster mitigation strategies. Manual mapping techniques are laborious and time-consuming, and thus not ideal in rapid response situations. Automated landslide mapping from optical satellite imagery using deep neural networks (DNNs) is becoming popular. However, distinguishing landslides from other changed objects in optical imagery using backbone DNNs alone is difficult. Attention modules have been introduced recently into the architecture of DNNs to address this problem by improving the discriminative ability of DNNs and suppressing noisy backgrounds. This study compares two state-of-the-art attention-boosted deep Siamese neural networks in mapping rainfall-induced landslides in the mountainous Himalayan region of Nepal using Planetscope (PS) satellite imagery. Our findings confirm that attention networks improve the performance of DNNs as they can extract more discriminative features. The Siamese Nested U-Net (SNUNet) produced the best and most coherent landslide inventory map among the methods in the test area, achieving an F1-score of 0.73, which is comparable to other similar studies. Our findings demonstrate a prospect for application of the attention-based DNNs in rapid landslide mapping and disaster mitigation not only for rainfall-triggered landslides but also for earthquake-triggered landslides.</p>}},
  author       = {{Amankwah, Solomon Obiri Yeboah and Wang, Guojie and Gnyawali, Kaushal and Hagan, Daniel Fiifi Tawiah and Sarfo, Isaac and Zhen, Dong and Nooni, Isaac Kwesi and Ullah, Waheed and Duan, Zheng}},
  issn         = {{1612-510X}},
  keywords     = {{Attention module; Change detection; Deep neural network (DNN); Landslide mapping}},
  language     = {{eng}},
  number       = {{10}},
  pages        = {{2459--2471}},
  publisher    = {{Springer}},
  series       = {{Landslides}},
  title        = {{Landslide detection from bitemporal satellite imagery using attention-based deep neural networks}},
  url          = {{http://dx.doi.org/10.1007/s10346-022-01915-6}},
  doi          = {{10.1007/s10346-022-01915-6}},
  volume       = {{19}},
  year         = {{2022}},
}