A Novel Method for Extracting Time Series Information of Deformation Area of a Single Landslide Based on Improved U-Net Neural Network

Bibo Dai; Bibo Dai; Bibo Dai; Yunmin Wang; Yunmin Wang; Chunyang Ye; Qihang Li; Canming Yuan; Song Lu; Song Lu; Yuyang Li

doi:10.3389/feart.2021.785476

Frontiers in Earth Science (Dec 2021)

A Novel Method for Extracting Time Series Information of Deformation Area of a Single Landslide Based on Improved U-Net Neural Network

Bibo Dai,
Bibo Dai,
Bibo Dai,
Yunmin Wang,
Yunmin Wang,
Chunyang Ye,
Qihang Li,
Canming Yuan,
Song Lu,
Song Lu,
Yuyang Li

Affiliations

Bibo Dai: School of Resources and Civil Engineering, Northeastern University, Shenyang, China
Bibo Dai: Geomechanics Research Center, Northeastern University, Shenyang, China
Bibo Dai: State Key Laboratory of Safety and Health for Metal Mines, Maanshan, China
Yunmin Wang: School of Resources and Environmental Engineering, Jiangxi University of Science and Technology, Ganzhou, China
Yunmin Wang: State Key Laboratory of Safety and Health for Metal Mines, Maanshan, China
Chunyang Ye: School of Resources and Environmental Engineering, Jiangxi University of Science and Technology, Ganzhou, China
Qihang Li: School of Resources and Environmental Engineering, Jiangxi University of Science and Technology, Ganzhou, China
Canming Yuan: School of Resources and Environmental Engineering, Jiangxi University of Science and Technology, Ganzhou, China
Song Lu: Quanzhou Institute of Equipment Manufacturing, Haixi Institutes, Chinese Academy of Sciences, Quanzhou, China
Song Lu: Zhejiang Zhipu Engineering Technology Corporation Limited, Huzhou, China
Yuyang Li: School of Resources and Environmental Engineering, Jiangxi University of Science and Technology, Ganzhou, China

DOI: https://doi.org/10.3389/feart.2021.785476
Journal volume & issue: Vol. 9

Abstract

Read online

This paper proposed an improved U-Net fully convolutional neural network to automatically extract a single landslide deformation information under time series based on the physical model experiments. This method extracts time series information for three different landslide deformation ranges. Compared to U-Net and mainstream superpixel method, evaluation indicators of DSC, VOE and RVD verify the high recognition accuracy and strong robustness of our method.

Published in Frontiers in Earth Science

ISSN: 2296-6463 (Online)
Publisher: Frontiers Media S.A.
Country of publisher: Switzerland
LCC subjects: Science
Website: https://www.frontiersin.org/journals/earth-science

About the journal

Abstract

Keywords