«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《地球科学与环境学报》[ISSN:1672-6561/CN:61-1423/P]

卷:

第44卷

期数:

2022年第03期

页码:

568-579

栏目:

大地测量、遥感与地学大数据

出版日期:

2022-05-15

文章信息/Info

Title:

Convolutional Neural Network Landslide Recognition Based on Terrain Feature Fusion

文章编号:

1672-6561(2022)03-0568-12

作者:

蔡浩杰¹; 2; 韩海辉¹; 2; 张雨莲¹; 王立社¹

(1. 中国地质调查局西安地质调查中心,陕西 西安 710054; 2. 中国遥感应用协会黄河流域高质量发展遥感分会,陕西 西安 710054)

Author(s):

CAI Hao-jie¹; 2;  HAN Hai-hui¹; 2;  ZHANG Yu-lian¹;  WANG Li-she¹

(1. Xi'an Center of Geological Survey, China Geological Survey, Xi'an 710054, Shaanxi, China; 2. Remote Sensing Application Branch for High-quality Development of the Yellow River Basin, China Association of Remote Sensing Application, Xi'an 710054, Shaanxi, China)

关键词:

地质灾害;  滑坡识别;  卷积神经网络;  遥感图像;  地形因子;  深度学习;  特征融合;  四川

Keywords:

geological hazard;  landslide recognition;  convolutional neural network;  remote sensing image;  terrain factor;  deep learning;  feature fusion;  Sichuan

分类号:

P642.22; TP183

DOI:

10.19814/j.jese.2021.12016

文献标志码:

A

摘要:

滑坡严重威胁着人民群众的生命财产安全。完整、准确的滑坡编录图是研究滑坡的重要资料。深度卷积神经网络方法由于众多优势而备受关注,然而卷积神经网络结构复杂,需要大量的训练样本,制约了其在滑坡制图上的发展。提出了融合地形特征的卷积神经网络建模方法。首先在遥感影像上叠加地形因子构建新的滑坡样本,然后设计提取并融合空间与光谱特征的轻量级卷积神经网络(FF-CNN),最后训练最优模型进行滑坡识别。在四川汶川地区进行的消融实验证明:在空间特征基础上融合光谱特征的FF-CNN模型滑坡识别评价指标F1分数和平均交并比(MIoU)分别提高0.020 2和0.014 4; 在遥感影像上叠加地形因子后,FF-CNN模型滑坡识别评价指标F1分数和MIoU值分别提高0.066 4和0.048 2。在湖北省三峡库区和四川省都江堰市虹口乡的实验说明FF-CNN模型表现出较强的适用性和迁移能力,在滑坡识别上具有较大潜力。

Abstract:

Landslides seriously threaten the safety of people's lives and property. A complete and accurate landslide inventory map is important for the study on landslides. The deep convolutional neural network method has attracted great attention due to its numerous advantages. However, the convolutional neural network has a complex structure and requires lots of training samples, which restrict the development of this technology in landslide recognition. A convolutional neural network modeling method incorporating terrain features was proposed. Firstly, a new landslide sample is constructed by superimposing terrain factors on remote sensing images, then a lightweight convolutional neural network that extracts and fuses spatial and spectral features(FF-CNN)is designed, and finally the optimal model is trained for landslide recognition. The ablation experiments in Wenchuan area of Sichuan show that after fusing spectral features on the basis of spatial features, the F1 score and MIoU of FF-CNN model are increased by 0.020 2 and 0.014 4, respectively; after superimposing the terrain factor on the remote sensing image, the F1 score and MIoU of FF-CNN model are increased by 0.066 4 and 0.048 2, respectively. The experiments in Three Gorges reservoir area of Hubei province and Hongkou town of Dujiangyan city, Sichuan province, show that FF-CNN model has strong applicability, migration ability and great potential in landslide recognition.

参考文献/References:

[1] 李郎平,兰恒星.滑坡运动路径复杂度研究:综述与展望[J].地球科学,2021,DOI:10.3799/dqkx.2021.224.
LI Lang-ping,LAN Heng-xing.Complexities of Landslide Moving Path:A Review and Perspective[J].Earth Science,2021,DOI:10.3799/dqkx.2021.224.
[2] CHEN Z Y,MENG X M,YIN Y P,et al.Landslide Research in China[J].Quarterly Journal of Engineering Geology and Hydrogeology,2016,49(4):279-285.
[3] 杨成生,董继红,朱赛楠,等.金沙江结合带巴塘段滑坡群InSAR探测识别与形变特征[J].地球科学与环境学报,2021,43(2):398-408.
YANG Cheng-sheng,DONG Ji-hong,ZHU Sai-nan,et al.Detection,Identification and Deformation Cha-racteristics of Landslide Groups by InSAR in Batang Section of Jinsha River Covergence Zone,China[J].Journal of Earth Sciences and Environment,2021,43(2):398-408.
[4] 自然资源部地质灾害技术指导中心.全国地质灾害通报(2021年)[R].北京:中国地质环境监测院,2022.
Geological Hazard Technology Instruction Center,Mi-nistry of Natural Resources.National Notification of Geological Disasters(2021)[R]. Beijing:China Institute of Geological Environment Monitoring,2022.
[5] MUSTAFA M,BISWAJEET P,HOSSEIN R.Improving Landslide Detection from Airborne Laser Scanning Data Using Optimized Dempster-Shafer[J].Remote Sensing,2018,10(7):1029.
[6] PRAKASH N,MANCONI A,LOEW S.Mapping Landslides on EO Data:Performance of Deep Learning Models vs.Traditional Machine Learning Models[J].Remote Sensing,2020,12(3):346.
[7] GUZZETTI F,MONDINI A C,CARDINALI M,et al.Landslide Inventory Maps:New Tools for an Old Pro-blem[J].Earth-science Reviews,2012,112(1/2):42-66.
[8] ZHU M,HE Y Q,HE Q Y.A Review of Researches on Deep Learning in Remote Sensing Application[J].International Journal of Geosciences,2019,10(1):1-11.
[9] 李德仁,童庆禧,李荣兴,等.高分辨率对地观测的若干前沿科学问题[J].中国科学:地球科学,2012,42(6):805-813.
LI De-ren,TONG Qing-xi,LI Rong-xing,et al.Current Issues in High-resolution Earth Observation Te-chnology[J].Science China:Earth Sciences,2012,42(6):805-813.
[10] HE F,TAN S C,LIU H J.Mechanism of Rainfall Induced Landslides in Yunnan Province Using Multi-scale Spatiotemporal Analysis and Remote Sensing Interpretation[J].Microprocessors and Microsystems,2022,90:104502.
[11] CHEN Y Y,MING D P,LING X,et al.Landslide Susceptibility Mapping Using Feature Fusion-based CPCNN-ML in Lantau Island,Hong Kong[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:3625-3639.
[12] 苏凤环,刘洪江,韩用顺.汶川地震山地灾害遥感快速提取及其分布特点分析[J].遥感学报,2008,12(6):956-963.
SU Feng-huan,LIU Hong-jiang,HAN Yong-shun.The Extraction of Mountain Hazard Induced by Wenchuan Earthquake and Analysis of Its Distributing Characteristic[J].Journal of Remote Sensing,2008,12(6):956-963.
[13] 陆会燕,李为乐,许 强,等.光学遥感与InSAR结合的金沙江白格滑坡上下游滑坡隐患早期识别[J].武汉大学学报(信息科学版),2019,44(9):1342-1354.
LU Hui-yan,LI Wei-le,XU Qiang,et al.Early Detection of Landslides in the Upstream and Downstream Areas of the Baige Landslide,the Jinsha River Based on Optical Remote Sensing and InSAR Technologies[J].Geomatics and Information Science of Wuhan University,2019,44(9):1342-1354.
[14] 姚纪华,吕慧珠,赵文刚,等.基于遥感与地质耦合法的水库滑坡群特性解译研究[J].工程勘察,2021,49(1):47-53.
YAO Ji-hua,LU Hui-zhu,ZHAO Wen-gang,et al.Interpretation of Characteristics of Reservoir Landslides Based on Remote Sensing and Geological Coupling Method[J].Geotechnical Investigation and Surveying,2021,49(1):47-53.
[15] BAI S B,WANG J,LÜ G N,et al.GIS-based Logistic Regression for Landslide Susceptibility Mapping of the Zhongxian Segment in the Three Gorges Area,China[J].Geomorphology,2010,115(1/2):23-31.
[16] KAVZOGLU T,COLKESEN I,SAHIN E K.Machine Learning Techniques in Landslide Susceptibility Mapping:A Survey and a Case Study[M]∥PRADHAN S P,VISHAL V,SINGH T N.Landslides:Theory,Practice and Modelling.Berlin:Springer,2019:283-301.
[17] 陈善静,向朝参,康 青,等.基于多源遥感时空谱特征融合的滑坡灾害检测方法[J].计算机研究与发展,2020,57(9):1877-1887.
CHEN Shan-jing,XIANG Chao-can,KANG Qing,et al.Multi-source Remote Sensing Based Accurate Landslide Detection Leveraging Spatial-temporal-spectral Feature Fusion[J].Journal of Computer Research and Development,2020,57(9):1877-1887.
[18] GOETZ J N,BRENNING A,PETSCHKO H,et al.Evaluating Machine Learning and Statistical Prediction Techniques for Landslide Susceptibility Modeling[J].Computers and Geosciences,2015,81:1-11.
[19] PHAM,B T,BUI D T,et al.Rotation Forest Fuzzy Rule-based Classifier Ensemble for Spatial Prediction of Landslides Using GIS[J].Natural Hazards,2016,83:97-127.
[20] BRENNING A.Spatial Prediction Models for Landslide Hazards:Review,Comparison and Evaluation[J].Natural Hazards and Earth System Sciences,2005,5(6):853-862.
[21] CONSTANTIN M,BEDNARIK M,JURCHESCU M C,et al.Landslide Susceptibility Assessment Using the Bivariate Statistical Analysis and the Index of Entropy in the Sibiciu Basin(Romania)[J].Environmental Earth Sciences,2011,63(2):397-406.
[22] ROBIULHOSSAIN M,DU Q,ALI C G,et al.Attention-based Domain Adaptation Using Residual Network for Hyperspectral Image Classification[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2020,13:6424-6433.
[23] TIAN Y,LI Z,LIN Y W,et al.Metal Object Detection for Electric Vehicle Inductive Power Transfer Systems Based On Hyperspectral Imaging[J].Mea-surement,2021,168:108493.
[24] RICHARDSON F,REYNOLDS D,DEHAK N.Deep Neural Network Approaches to Speaker and Language Recognition[J].IEEE Signal Processing Letters,2015,22(10):1671-1675.
[25] LU H,MA L,FU X,et al.Landslides Information Extraction Using Object-oriented Image Analysis Paradigm Based on Deep Learning and Transfer Learning[J].Remote Sensing,2020,12(5):752.
[26] 巨袁臻,许 强,金时超,等.使用深度学习方法实现黄土滑坡自动识别[J].武汉大学学报(信息科学版),2020,45(11):1747-1755.
JU Yuan-zhen,XU Qiang,JIN Shi-chao,et al.Automatic Object Detection of Loess Landslide Based on Deep Learning[J].Geomatics and Information Science of Wuhan University,2020,45(11):1747-1755.
[27] GAO X,CHEN T,NIU R Q,et al.Recognition and Mapping of Landslide Using a Fully Convolutional DenseNet and Influencing Factors[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:7881-7894.
[28] LIU T,CHEN T,NIU R Q,et al.Landslide Detection Mapping Employing CNN,ResNet,and DenseNet in the Three Gorges Reservoir,China[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:11417-11428.
[29] LECUN Y,BENGIO Y,HINTON G.Deep Learning[J].Nature,2015,521:436-444.
[30] GHORBANZADEH O,BLASCHKE T,GHOLAMNIA K,et al.Evaluation of Different Machine Learning Methods and Deep-learning Convolutional Neural Networks for Landslide Detection[J].Remote Sensing,2019,11(2):196.
[31] PIKE R J.The Geometric Signature:Quantifying Landslide-terrain Types from Digital Elevation Models[J].Mathematical Geology,1988,20(5):491-511.
[32] JABOYEDOFF M,OPPIKOFER T,ABELLÁN A,et al.Use of LIDAR in Landslide Investigations:A Review[J].Natural Hazards,2012,61:5-28.
[33] LESHCHINSKY B A,OLSEN M J,TANYU B F.Contour Connection Method for Automated Identification and Classification of Landslide Deposits[J].Computers and Geosciences,2015,74:27-38.
[34] LI G,WEST A J,DENSMORE A L,et al.Seismic Mountain Building:Landslides Associated with the 2008 Wenchuan Earthquake in the Context of a Ge-neralized Model for Earthquake Volume Balance[J].Geochemistry,Geophysics,Geosystems,2014,15(4):833-844.
[35] DASH M,LIU H.Feature Selection for Classification[J].Intelligent Data Analysis,1997,1(1/2/3/4):131-156.
[36] LECUN Y,BOTTOU L,BENGIO Y,et al.Gradient-based Learning Applied to Document Recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[37] XU X D,LI W,RAN Q,et al.Multisource Remote Sensing Data Classification Based on Convolutional Neural Network[J].IEEE Transactions on Geoscience and Remote Sensing,2018,56(2):937-949.
[38] DONG L,WEI F,ZHOU M,et al.Question Answering over Freebase with Multi-column Convolutional Neural Networks[C]∥Association for Computational Linguistics.Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing.Beijing:Association for Computational Linguistics,2015:260-269.
[39] CHEN T,NIU R,JIA X.A Comparison of Information Value and Logistic Regression Models in Landslide Susceptibility Mapping by Using GIS[J].Environmental Earth Sciences,2016,75(10):861-867.
[40] CAI H J,CHEN T,NIU R Q,et al.Landslide Detection Using Densely Connected Convolutional Networks and Environmental Conditions[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:5235-5247.

相似文献/References:

[1]王运生,全清,罗永红,等.四川芦山Ms7.0级地震的地质环境影响分析[J].地球科学与环境学报,2013,35(02):92.
　WANG Yun-sheng,QUAN Qing,LUO Yong-hong,et al.Geological Environmental Influence of Lushan Ms7.0 Earthquake in Sichuan[J].Journal of Earth Sciences and Environment,2013,35(03):92.
[2]宋彦辉,李忠生,高虎艳,等.西安三类勘察场地隐伏地裂缝识别特征[J].地球科学与环境学报,2015,37(04):94.
　SONG Yan-hui,LI Zhong-sheng,GAO Hu-yan,et al.Recognition Features of Non-outcropping Ground Fissures for the Third Investigation Site in Xi’an[J].Journal of Earth Sciences and Environment,2015,37(03):94.
[3]彭建兵,马润勇,范 文,等.汶川大震的科学思考[J].地球科学与环境学报,2009,31(01):1.
　PENG Jian-bing,MA Run-yong,FAN Wen,et al.Science Contemplation for Wenchuan Earthquake of 12 May, 2008[J].Journal of Earth Sciences and Environment,2009,31(03):1.
[4]邓亚虹,彭建兵,卢全中,等.地铁工程地质灾害危险性综合评估定量方法——以西安地铁一号线为例[J].地球科学与环境学报,2009,31(03):291.
　DENG Ya-hong,PENG Jian-bing,LU Quan-zhong,et al.Quantification Grading Method in Subway Engineering Risk Assessment of Geological Disasters——Taking No.1 Subway in Xi'an as an Example[J].Journal of Earth Sciences and Environment,2009,31(03):291.
[5]马润勇,彭建兵,袁志东,等.青藏高原隆升的黄土高原构造侵蚀效应[J].地球科学与环境学报,2007,29(03):289.
　MA Run-yong,PENG Jian-bing,YUAN Zhi-dong,et al.Geological Hazard Effect in Loess Plateau due to Qinghai-Tibet Plateau Uplift[J].Journal of Earth Sciences and Environment,2007,29(03):289.
[6]胡建刚,刘勇平,姜海波.怒江左岸小型水电工程地质环境适宜性评价[J].地球科学与环境学报,2007,29(03):294.
　HU Jian-gang,LIU Yong-ping,JIANG Hai-bo.Geological Environment Feasibility for Constructing Small-Scale Hydropower Engineering on Nujiang River Left Bank[J].Journal of Earth Sciences and Environment,2007,29(03):294.
[7]田梅青,姜振泉,刘炜金,等.山东省栖霞市泥石流灾害的 成生环境及危险性趋势分区评价[J].地球科学与环境学报,2005,27(04):24.
　TIAN Mei-qing,JIANG Zhen-quan,LIU Wei-jin,et al.Zonal Evaluation on Formation Environment and Dangerous Trend of Debris Flow in Qixia City of Shandong Province[J].Journal of Earth Sciences and Environment,2005,27(03):24.
[8]朱良峰,吴信才,殷坤龙,等.基于信息量模型的中国滑坡灾害风险区划研究[J].地球科学与环境学报,2004,26(03):52.
　ZHU Liang-feng,WU Xin-cai,YIN Kun-long,et al.Risk zonation of landslide in China based on information content model[J].Journal of Earth Sciences and Environment,2004,26(03):52.
[9]陈练武.耀州区地质灾害分区特征及防治对策[J].地球科学与环境学报,2004,26(03):74.
　CHEN Lian-wu.Characteristics and prevention countermeasures of geological hazards zones in Yaozhou district[J].Journal of Earth Sciences and Environment,2004,26(03):74.
[10]李守定,白亚恒,姜越,等.基于内外动力耦合成因理论的新疆地质灾害气象预警显式统计模型[J].地球科学与环境学报,2017,39(02):286.
　LI Shou-ding,BAI Ya-heng,JIANG Yue,et al.Explicit Statistic Meteorological Early-warning Model of Geological Hazards in Xinjiang, China Based on the Genesis Theory of Endogenic and Exogenic Coupling[J].Journal of Earth Sciences and Environment,2017,39(03):286.
[11]辛鲁斌,韩 玲*,李良志.基于多源数据融合的滑坡智能识别[J].地球科学与环境学报,2023,45(04):920.
　XIN Lu-bin,HAN Ling*,LI Liang-zhi.Landslide Intelligent Recognition Based on Multi-source Data Fusion[J].Journal of Earth Sciences and Environment,2023,45(03):920.

备注/Memo

备注/Memo:

收稿日期:2021-12-07; 修回日期:2022-03-23
基金项目:中国地质调查局地质调查项目(DD20211387,DD20211393)
作者简介:蔡浩杰(1996-),男,陕西宝鸡人,中国地质调查局西安地质调查中心助理工程师,E-mail:cason@cug.edu.cn。

常用功能

更新日期/Last Update: 2022-06-01