Please wait a minute...
img

官方微信

遥感技术与应用  2023, Vol. 38 Issue (5): 1180-1191    DOI: 10.11873/j.issn.1004-0323.2023.5.1180
遥感应用     
基于马尔科夫随机场的PLANET高分影像滑坡提取研究
高梦洁1,2,3(),陈方1,2,3(),王雷1,2,杨阿强1,2,于博1,2
1.中国科学院空天信息创新研究院,中国科学院数字地球重点实验室,北京 100094
2.可持续发展大数据国际研究中心,北京 100094
3.中国科学院大学,资源与环境学院,北京 100049
Research on Extraction of Landslide from PLANET High Spatial Resolution Remote Sensing Image based on Markov Random Field
Mengjie GAO1,2,3(),Bo YU1,2,3(),Lei WANG1,2,Aqiang YANG1,2,Fang CHEN1,2
1.Key Laboratory of Digital Earth Science,Aerospace Information Research Institute,Chinese Academy of Sciences,No. 9 Dengzhuang South Road,Beijing 100094,China
2.International Research Center of Big Data for Sustainable Development Goals,Beijing 100094,China
3.University of Chinese Academy of Sciences,Beijing 100049,China
 全文: PDF(12167 KB)   HTML
摘要:

滑坡是自然界频繁发生的地质灾害,会对人民的生命造成威胁,并带来巨大的财产损失。因此,高效准确地进行滑坡提取对快速制定应急救灾方案、减少损失有着重要的意义。目前滑坡提取研究大多针对单个或少许事件,且背景地物比较单一。研究采用高分辨率遥感影像,针对复杂背景地物条件下的多起滑坡构建提取模型,并对其进行精度验证。采用马尔科夫随机场最小化能量方程结果作为滑坡提取特征,并用于滑坡提取模型构建,与目前滑坡提取研究中常用的特征相比较,验证这一特征对于滑坡提取的有效性。选用多时相的PLANET 3 m分辨率遥感影像,对2018年9月6日北海道地震引发的多起滑坡进行提取。结果表明:本研究提出的特征运用于滑坡提取中,可以提高提取精度2%,在滑坡提取的完整性上得到一定提升,为在大区域范围内的滑坡精确提取提供帮助。

关键词: 滑坡提取随机森林变化检测马尔科夫随机场    
Abstract:

Landslides represent a prevalent geological hazard with serious consequences for the safety of human lives and properties. Therefore, the efficient and accurate extraction of landslides is of great significance for the timely development of emergency response plans aimed at reducing losses. Current studies typically focus on single or a few events, often under relatively simple background conditions. To address these limitations, we propose using high-resolution remote sensing images to extract multiple landslides under complex background conditions, with the precision of the approach being verified. Specifically, we construct a landslide extraction model that utilizes the result of the minimization of energy equation using Markov random field as a feature. To evaluate the effectiveness of the model, we compare it with features commonly used in landslide extraction research. We select multi-temporal Planet 3M resolution remote sensing images to extract landslides in Hokkaido on September 6, 2018. Our results demonstrate that the proposed feature improves the accuracy of landslide extraction by 2%, while also improving the integrity of the extraction to a certain extent. This approach offers valuable assistance for accurately extracting landslides in large areas.

Key words: Landslide detection    Random forest    Change detection    Markov random field
收稿日期: 2022-07-06 出版日期: 2023-11-07
ZTFLH:  P642.22  
基金资助: 中国—东盟地球大数据平台与应用示范项目(桂科AA20302022)
通讯作者: 陈方     E-mail: gaomengjie21@mails.ucas.ac.cn;chenfang_group@radi.ac.cn
作者简介: 高梦洁(1999-),女,湖北武汉人,博士研究生,主要从事基于遥感影像的信息提取研究。E?mail:gaomengjie21@mails.ucas.ac.cn
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
高梦洁
陈方
王雷
杨阿强
于博

引用本文:

高梦洁,陈方,王雷,杨阿强,于博. 基于马尔科夫随机场的PLANET高分影像滑坡提取研究[J]. 遥感技术与应用, 2023, 38(5): 1180-1191.

Mengjie GAO,Bo YU,Lei WANG,Aqiang YANG,Fang CHEN. Research on Extraction of Landslide from PLANET High Spatial Resolution Remote Sensing Image based on Markov Random Field. Remote Sensing Technology and Application, 2023, 38(5): 1180-1191.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2023.5.1180        http://www.rsta.ac.cn/CN/Y2023/V38/I5/1180

图1  研究区图
图2  滑坡影像图
参数详细描述
卫星数量100余颗
运行轨道太阳同步轨道(475~600 km)
国际空间站轨道(约400 km)
分辨率3 m
光谱波段Band1:蓝(455~515 nm)
Band2:绿(500~590 nm)
Band3:红(590~670 nm)
Band4:近红外(780~860 nm)
产品级别1B:基础的产品数据,已进行传感器和辐射定标
3B:已经进行过传感器、辐射、正射校正以及大气校正的正射产品
表1  PLANET数据参数
图3  滑坡位置图
图4  技术流程图
编号特征类型特征描述
1光谱特征红色波段
2绿色波段
3蓝色波段
4近红外波段
5纹理特征GLCM对比度
6GLCM自相关
7GLCM ASM能量
8GLCM相异性
9GLCM同质性
10GLCM能量
11地形特征DEM
表2  随机森林的特征选择
图5  训练数据与测试数据分布图
图6  滑坡发生前后影像图
图7  结合马尔科夫随机场的随机森林滑坡提取结果图
图像区域精度/%召回率/%F1值
174.068.90.71
277.775.40.77
379.174.90.77
471.172.10.72
表3  改进模型提取精度
图8  不采用改进特征的随机森林对比实验结果图
图像区域精度/%召回率/%F1值
169.870.20.70
276.576.30.76
387.375.60.81
468.479.30.73
表4  不采用改进特征的模型提取精度
特征重要性
本文提出的改进特征0.231
DEM0.205
蓝色波段0.169
近红外波段0.150
绿色波段0.090
红色波段0.079
GLCM相异性0.019
GLCM能量0.019
GLCM同质性0.016
GLCM自相关0.009
GLCM ASM能量0.007
GLCM对比度0.006
表5  随机森林中各特征的重要性
图9  滑坡提取结果对比图
图10  被误分为滑坡的裸地
1 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).DOI:10.3390/rs11020196
doi: 10.3390/rs11020196
2 ZHANG Y, WU W, QIN Y, et al. Mapping landslide hazard risk using random forest algorithm in Guixi, Jiangxi, China [J]. ISPRS International Journal of Geo-Information,2020,9(11):695.DOI:10.3390/ijgi9110695
doi: 10.3390/ijgi9110695
3 LU P, QIN Y, LI Z, et al. Landslide mapping from multi-sensor data through improved change detection-based markov random field[J]. Remote Sensing of Environment, 2019, 231:111235.
4 ZHANG Haitao. Key techniques study of landslides extraction based on high-resolution remote sensing images[D]. Wuhan:China University of Geoscience, 2017.
4 张海涛. 基于高分影像的滑坡提取关键技术研究[D]. 武汉:中国地质大学(武汉), 2017.
5 LI Deren. Towards photogrammetry and remote sensing: Status and future development[J]. Journal of Wuhan Technical University of Surveying and Mapping,2000,25(1):1-6.
5 李德仁. 摄影测量与遥感的现状及发展趋势[J]. 武汉测绘科技大学学报, 2000,25(1):1-6.
6 JIA Jutao. Automatic identification of landslide based on GF-1 satellite image—Taking Shuicheng county of Guizhou province as an example[D]. Mianyang:Southwest University of Science and Technology, 2020.
6 贾菊桃. 基于高分一号卫星影像的滑坡自动识别[D]. 绵阳:西南科技大学, 2020.
7 WANG Ning, CHEN Fang, YU Bo. A object-oriented landslide extraction method based on morphological opening operation[J]. Remote Sensing Technology and Application. 2018, 33(3): 520-529.
7 王宁, 陈方, 于博. 基于形态学开运算的面向对象滑坡提取方法研究[J]. 遥感技术与应用, 2018, 33(3): 520-529.
8 GAO Siyan, HUANG Shifeng, SUN Yayong, et al. Research on the application of sentinel-1 SAR data in monitoring of landslide disaster in Maoxian county Sichuan[J]. Journal of China Institute of Water Resources and Hydropower Research, 2019, 17(2): 132-8.
8 高思远, 黄诗峰, 孙亚勇,等. Sentinel-1 SAR数据在四川茂县山体滑坡灾害监测中的应用研究[J]. 中国水利水电科学研究院学报, 2019, 17(2): 132-138.
9 GORSEVSKI P V, BROWN M K, PANTER K, et al. Landslide detection and susceptibility mapping using LiDAR and an artificial neural network approach: A case study in the Cuyahoga Valley National Park,Ohio[J]. Landslides, 2016, 13(3): 467-484.
10 MONDINI A C, GUZZETTI F, REICHENBACH P, et al. Semi-automatic recognition and mapping of rainfall induced shallow landslides using optical satellite images[J]. Remote Sensing of Environment, 2011, 115(7): 1743-1757.
11 DANNEELS G, PIRARD E, H-B HAVENITH, et al. Automatic landslide detection from remote sensing images using supervised classification methods[C]∥ IGARSS: 2007 Ieee International Geoscience and Remote Sensing Symposium, Vols 1-12:Sensing and Understanding Our PlanET.2007: 3014-+. 10.1109/IGARSS.2007.4423479
doi: 10.1109/IGARSS.2007.4423479
12 STUMPF A, KERLE N. Object-oriented mapping of landslides using random forests [J]. Remote Sensing of Environment, 2011, 115(10): 2564-2577.
13 WANG Heping, YUAN Huijie, LIU Zhengjun, et al. Research on object-oriented landslide extraction method based on visible image[J]. Geospatial Information,2020, 18(2): 108-12+8.
13 王和平, 袁慧洁, 刘正军, 等. 基于可见光影像的面向对象滑坡提取方法研究[J]. 地理空间信息, 2020, 18(2): 108-12+8.
14 LU P, STUMPF A, KERLE N,et al. Object-oriented change detection for landslide rapid mapping[J]. IEEE Geoscience And Remote Sensing Letters, 2011, 8(4): 701-705.
15 SI A, ZHANG J, TONG S, et al. Regional landslide identification based on susceptibility analysis and change detection [J]. ISPRS International Journal of Geo-Information, 2018, 7(10):394.
16 XU Qiao, MENG Fanli, YU Shaohuai. Landslide remote sensing interpretation based on object-oriented random forest classification model[J]. Journal of China & Foreign Highway, 2019, 39(3): 30-34.
16 徐乔, 孟凡利, 余绍淮. 基于面向对象随机森林分类模型的滑坡遥感解译[J]. 中外公路, 2019, 39(3): 30-34.
17 CHEN W, LI X, WANG Y, et al. Forested landslide detection using lidar data and the random forest algorithm: A case study of the Three Gorges, China[J]. Remote Sensing Of Environment, 2014, 152: 291-301.
18 LIU P, WEI Y, WANG Q, et al. A research on landslides automatic extraction model based on the improved mask R-CNN[J]. ISPRS International Journal of Geo-Information, 2021, 10(3):168.
19 YI Y, ZHANG W. A new deep-learning-based approach for earthquake-triggered landslide detection from single-temporal rapideye satellite imagery [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 6166-6176.
20 FU Xiao, GUO Jiawei, LIU Xiuju, et al. Method of earthquake landslide information extraction based on high resolution unmanned aerial vehicle images[J]. Journal of Seism Ological Research, 2018, 41(2): 186-191.
20 付萧, 郭加伟, 刘秀菊, 等. 无人机高分辨率遥感影像地震滑坡信息提取方法[J]. 地震研究, 2018, 41(2): 186-191.
21 YU B, CHEN F, XU C. Landslide detection based on contour-based deep learning framework in case of national scale of Nepal in 2015[J]. Computers & Geosciences, 2020, 135:104388.
22 CHEN F, YU B, XU C, et al. Landslide detection using probability regression, A case study of Wenchuan, northwest of Chengdu [J]. Applied Geography, 2017, 89: 32-40.
23 SHAO X, MA S, XU C, et al. Planet image-based inventorying and machine learning-based susceptibility mapping for the landslides triggered by the 2018 mw6.6 Tomakomai,Japan earthquake [J]. Remote Sensing, 2019, 11(8):978.
24 YAMAGISHI H, YAMAZAKI F. Landslides by the 2018 Hokkaido Iburi-Tobu earthquake on september 6[J]. Landslides, 2018, 15(12): 2521-2524.
25 QIN Xuan. Improvement and application of random forest algorithm[D]. Lanzhou:Lanzhou Jiaotong University, 2020.
25 秦璇. 随机森林算法的改进及应用[D]. 兰州:兰州交通大学, 2020.
26 AKINCI H, KILICOGLU C, DOGAN S. Random forest-based landslide susceptibility mapping in coastal regions of Artvin,Turkey[J].Isprs International Journal of Geo-Information, 2020,9(9):553.
27 BAI Jie, JU Nengpan, ZHANG Chengqiang, et al. Characteristics and successful early warning case of Xingyi landslide in Guizhou province[J]. Journalof Engineering Geology, 2020, 28(6): 1246-1258.
27 白洁, 巨能攀, 张成强, 等. 贵州兴义滑坡特征及过程预警研究[J]. 工程地质学报, 2020, 28(6): 1246-1258.
28 LIU Guangyu, HUANG Yi, CAO Yu, et al. Research on extraction of image texture feature based on gray co-occurrence matrix[J]. Technology Wind, 2021, (12): 61-64.
28 刘光宇, 黄懿, 曹禹, 等. 基于灰度共生矩阵的图像纹理特征提取研究[J]. 科技风, 2021, (12): 61-64.
29 LI Z, SHI W, LU P, et al. Landslide mapping from aerial photographs using change detection-based Markov random field[J]. Remote Sensing of Environment,2016,187:76-90.
30 MENZE B H, KELM B M, MASUCH R, et al. A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data [J]. Bmc Bioinformatics, 2009, 10:1-16.
31 WAN S, LEI T C, MA H L, et al. The Analysis on similarity of spectrum analysis of landslide and bareland through hyper-spectrum image bands[J]. Water, 2019, 11(11):2414.
[1] 张成才,刘威,杨峰,彭凯,周雪丽. 基于编码对齐生成对抗网络的异源遥感影像地表覆被变化检测方法[J]. 遥感技术与应用, 2023, 38(5): 1107-1117.
[2] 杜懿,林泽群,庄胜杰,陈润庭,王大刚. GPM卫星降水产品在长江流域的空间降尺度研究[J]. 遥感技术与应用, 2023, 38(3): 697-707.
[3] 张丽坤,潘一凡,赵楚雯,邱国梁,周佩,陈翔,汪洋. 基于随机森林算法的华北平原灰霾区域识别研究[J]. 遥感技术与应用, 2023, 38(3): 752-766.
[4] 匡开新,杨英宝,高永年,刘宇翔. 协同Sentinel合成孔径雷达和光学影像多特征的不透水面随机森林提取方法[J]. 遥感技术与应用, 2023, 38(2): 422-431.
[5] 范霄,孔金玲,钟炎伶,蒋镒竹,张静雅. 基于XGBoost算法的遥感图像云检测[J]. 遥感技术与应用, 2023, 38(1): 156-162.
[6] 张妮娜,张珂,李运平,李曦,刘涛. 中国南方典型湿润山区植被类型的无人机多光谱遥感机器学习分类研究[J]. 遥感技术与应用, 2023, 38(1): 163-172.
[7] 凌菊,李爱农,靳华安. 基于DART模型和随机森林估算山地叶面积指数[J]. 遥感技术与应用, 2023, 38(1): 39-50.
[8] 杜绍杰,赵天杰,施建成,马春锋,邹德富,王振,姚盼盼,彭志晴,郑景耀. Sentinel-1和Sentinel-2协同反演地表土壤水分[J]. 遥感技术与应用, 2022, 37(6): 1404-1413.
[9] 马超,黄华国,田昕,刘炳杰,温坤剑,王鹏杰. 基于背包激光雷达的单木材积估测方法研究[J]. 遥感技术与应用, 2022, 37(5): 1071-1083.
[10] 高帅,侯学会,汪云,王倩,陈悦,邢瑞,王晶. 基于机器学习和大数据平台的陆地生态系统碳收支遥感监测[J]. 遥感技术与应用, 2022, 37(5): 1190-1197.
[11] 张琪,张光辉,张岩,王佳希,余双武. 基于不同分辨率遥感影像自动提取切沟的精度分析和转换模型[J]. 遥感技术与应用, 2022, 37(5): 1217-1226.
[12] 陈绎冰,李天依,李欣艳,赵文川,范荣峰,陈凤娇,杨元建. 基于随机森林和遥感的台风降水云光谱与降水关系研究[J]. 遥感技术与应用, 2022, 37(5): 1277-1288.
[13] 韦中晖,靳海亮,顾晓鹤,杨英茹,王庚泽,潘瑜春. 基于多时相协同变化检测的耕地撂荒遥感监测[J]. 遥感技术与应用, 2022, 37(3): 539-549.
[14] 丁书培,李蒙蒙,汪小钦,李琳,吴瑞姣,黄姮. 基于时序遥感数据的福州市耕地非农化特征及驱动因子分析[J]. 遥感技术与应用, 2022, 37(3): 550-563.
[15] 李媛,周小成,陈芸芝,王锋克. 结合Landsat时序遥感干扰检测的福建将乐县森林林龄估算与评价[J]. 遥感技术与应用, 2022, 37(3): 651-662.