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

作     者：Chang-dong Li Peng-fei Feng Xi-hui Jiang Shuang Zhang Jie Meng Bing-chen Li Chang-dong Li;Peng-fei Feng;Xi-hui Jiang;Shuang Zhang;Jie Meng;Bing-chen Li

作者机构：Faculty of EngineeringChina University of GeoscienceWuhan 430074China Badong National Observation and Research Station of GeohazardsChina University of GeosciencesWuhan 430074China School of Mechanical Engineering and Electronic InformationChina University of GeosciencesWuhan 430074China College of Geology Engineering and GeomaticsChang’an UniversityXi’an 710054China 

出 版 物：《China Geology》 (中国地质（英文）)

年 卷 期：2024年第7卷第2期

页      面：277-290页

核心收录：

学科分类：083002[工学-环境工程] 0830[工学-环境科学与工程（可授工学、理学、农学学位）] 081802[工学-地球探测与信息技术] 081803[工学-地质工程] 08[工学] 0818[工学-地质资源与地质工程] 081602[工学-摄影测量与遥感] 0816[工学-测绘科学与技术] 

基　　金：supported by the National Natural Science Foundation of China(Grant Nos.42090054,41931295) the Natural Science Foundation of Hubei Province of China(2022CFA002) 

主　　题：Geohazard Landslide boundary detection Remote sensing image Deep learning model Steep slope Large annual rainfall Human settlements Infrastructure Agricultural land Eastern Tibetan Plateau Geological hazards survey engineering 

摘      要：The frequent occurrence of extreme weather events has rendered numerous landslides to a global natural disaster *** is crucial to rapidly and accurately determine the boundaries of landslides for geohazards evaluation and emergency ***,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 *** 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 *** 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 *** experimental results demonstrate that SCDnn can identify landslide boundaries in 119 images with MIoU values between 0.8and 0.9;while 52 images with MIoU values exceeding 0.9,which exceeds the identification accuracy of existing *** 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 inve stigations and applications in related domains.