Extensive identification of landslide boundaries using remote sensing images and deep learning method

ABSTRACT: The frequent occurrence of extreme weather events has rendered numerous landslides to a global natural disaster issue. It is crucial to rapidly and accurately determine the boundaries of landslides for geohazards evaluation and emergency response. Therefore, the Skip Connection DeepLab neu...

Full description

Saved in:
Bibliographic Details
Main Authors: Chang-dong Li, Peng-fei Feng, Xi-hui Jiang, Shuang Zhang, Jie Meng, Bing-chen Li
Format: Article
Language:English
Published: KeAi Communications Co., Ltd. 2024-04-01
Series:China Geology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2096519224001095
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849426198137929728
author Chang-dong Li
Peng-fei Feng
Xi-hui Jiang
Shuang Zhang
Jie Meng
Bing-chen Li
author_facet Chang-dong Li
Peng-fei Feng
Xi-hui Jiang
Shuang Zhang
Jie Meng
Bing-chen Li
author_sort Chang-dong Li
collection DOAJ
description ABSTRACT: The frequent occurrence of extreme weather events has rendered numerous landslides to a global natural disaster issue. It is crucial to rapidly and accurately determine the boundaries of landslides for geohazards evaluation and emergency response. Therefore, the Skip Connection DeepLab neural network (SCDnn), a deep learning model based on 770 optical remote sensing images of landslide, is proposed to improve the accuracy of landslide boundary detection. The SCDnn model is optimized for the over-segmentation issue which occurs in conventional deep learning models when there is a significant degree of similarity between topographical geomorphic features. SCDnn exhibits notable improvements in landslide feature extraction and semantic segmentation by combining an enhanced Atrous Spatial Pyramid Convolutional Block (ASPC) with a coding structure that reduces model complexity. The experimental results demonstrate that SCDnn can identify landslide boundaries in 119 images with MIoU values between 0.8 and 0.9; while 52 images with MIoU values exceeding 0.9, which exceeds the identification accuracy of existing techniques. This work can offer a novel technique for the automatic extensive identification of landslide boundaries in remote sensing images in addition to establishing the groundwork for future investigations and applications in related domains.
format Article
id doaj-art-4aabebc92abc4ea58be8f12666d4eaa4
institution Kabale University
issn 2589-9430
language English
publishDate 2024-04-01
publisher KeAi Communications Co., Ltd.
record_format Article
series China Geology
spelling doaj-art-4aabebc92abc4ea58be8f12666d4eaa42025-08-20T03:29:31ZengKeAi Communications Co., Ltd.China Geology2589-94302024-04-017227729010.31035/cg2023148Extensive identification of landslide boundaries using remote sensing images and deep learning methodChang-dong Li0Peng-fei Feng1Xi-hui Jiang2Shuang Zhang3Jie Meng4Bing-chen Li5Faculty of Engineering, China University of Geoscience, Wuhan 430074, China; Badong National Observation and Research Station of Geohazards, China University of Geosciences, Wuhan 430074, China; Corresponding author:School of Mechanical Engineering and Electronic Information, China University of Geosciences, Wuhan 430074, ChinaFaculty of Engineering, China University of Geoscience, Wuhan 430074, ChinaCollege of Geology Engineering and Geomatics, Chang'an University, Xi'an 710054, ChinaFaculty of Engineering, China University of Geoscience, Wuhan 430074, ChinaFaculty of Engineering, China University of Geoscience, Wuhan 430074, ChinaABSTRACT: The frequent occurrence of extreme weather events has rendered numerous landslides to a global natural disaster issue. It is crucial to rapidly and accurately determine the boundaries of landslides for geohazards evaluation and emergency response. Therefore, the Skip Connection DeepLab neural network (SCDnn), a deep learning model based on 770 optical remote sensing images of landslide, is proposed to improve the accuracy of landslide boundary detection. The SCDnn model is optimized for the over-segmentation issue which occurs in conventional deep learning models when there is a significant degree of similarity between topographical geomorphic features. SCDnn exhibits notable improvements in landslide feature extraction and semantic segmentation by combining an enhanced Atrous Spatial Pyramid Convolutional Block (ASPC) with a coding structure that reduces model complexity. The experimental results demonstrate that SCDnn can identify landslide boundaries in 119 images with MIoU values between 0.8 and 0.9; while 52 images with MIoU values exceeding 0.9, which exceeds the identification accuracy of existing techniques. This work can offer a novel technique for the automatic extensive identification of landslide boundaries in remote sensing images in addition to establishing the groundwork for future investigations and applications in related domains.http://www.sciencedirect.com/science/article/pii/S2096519224001095GeohazardLandslide boundary detectionRemote sensing imageDeep learning modelSteep slopeLarge annual rainfall
spellingShingle Chang-dong Li
Peng-fei Feng
Xi-hui Jiang
Shuang Zhang
Jie Meng
Bing-chen Li
Extensive identification of landslide boundaries using remote sensing images and deep learning method
China Geology
Geohazard
Landslide boundary detection
Remote sensing image
Deep learning model
Steep slope
Large annual rainfall
title Extensive identification of landslide boundaries using remote sensing images and deep learning method
title_full Extensive identification of landslide boundaries using remote sensing images and deep learning method
title_fullStr Extensive identification of landslide boundaries using remote sensing images and deep learning method
title_full_unstemmed Extensive identification of landslide boundaries using remote sensing images and deep learning method
title_short Extensive identification of landslide boundaries using remote sensing images and deep learning method
title_sort extensive identification of landslide boundaries using remote sensing images and deep learning method
topic Geohazard
Landslide boundary detection
Remote sensing image
Deep learning model
Steep slope
Large annual rainfall
url http://www.sciencedirect.com/science/article/pii/S2096519224001095
work_keys_str_mv AT changdongli extensiveidentificationoflandslideboundariesusingremotesensingimagesanddeeplearningmethod
AT pengfeifeng extensiveidentificationoflandslideboundariesusingremotesensingimagesanddeeplearningmethod
AT xihuijiang extensiveidentificationoflandslideboundariesusingremotesensingimagesanddeeplearningmethod
AT shuangzhang extensiveidentificationoflandslideboundariesusingremotesensingimagesanddeeplearningmethod
AT jiemeng extensiveidentificationoflandslideboundariesusingremotesensingimagesanddeeplearningmethod
AT bingchenli extensiveidentificationoflandslideboundariesusingremotesensingimagesanddeeplearningmethod