[1]王 鹏,葛 洁,方 峥,等.半自动面向对象高分遥感地灾目标提取方法[J].山地学报,2018,(04):654-659.[doi:10.16089/j.cnki.1008-2786.000361]
　WANG Peng,GE Jie,FANG Zheng,et al.Semi-automatic Object-oriented Geological Disaster Target Extraction based on High-resolution Remote Sensing[J].Mountain Research,2018,(04):654-659.[doi:10.16089/j.cnki.1008-2786.000361]
点击复制

半自动面向对象高分遥感地灾目标提取方法()
分享到：

《山地学报》[ISSN:1008-2186/CN:51-1516]

卷:
期数:
2018年04期
页码:
654-659
栏目:
山地技术
出版日期:
2018-07-30

文章信息/Info

Title:
Semi-automatic Object-oriented Geological Disaster Target Extraction based on High-resolution Remote Sensing
文章编号:
1008-2786-(2018)4-654-06
作者:
王 鹏1 ; 葛 洁1; 方 峥1; 赵国兵1; 孙根云2
1.四川省国土勘测规划研究院,成都 610045; 2.中国石油大学(华东)地球科学与技术学院,山东 青岛 266580
Author(s):
WANG Peng1;  GE Jie1;  FANG Zheng1;  ZHAO Guobing1;  SUN Genyun2
1.Sichuan Institute of Land Planning and Survey, Chengdu 610045, China; 2.School of Geosciences, China University of Petroleum, Qingdao 266580, Shandong, China
关键词:
地质灾害;  图像分割;  半自动;  高分辨率遥感
Keywords:
geological disasters;  image segmentation;  semi-automatic;  high-resolution remote sensing
分类号:
P237
DOI:
10.16089/j.cnki.1008-2786.000361
文献标志码:
A
摘要:
滑坡与泥石流等类型地质灾害发生后,快速准确的提取地质灾害信息对于灾害评估、救援具有重要意义。随着航空、航天技术的发展,具有宏观、快速、准确优势的遥感技术越来越多的应用于地灾分析。然而,在遥感影像上直接利用人工解译的工作方式费时费力,而计算机自动解译精度低。对此,本文提出一种半自动地灾目标提取方法。首先通过人机交互的方式进行简单的地灾目标标记,然后基于均值漂移算法进行图像分割,基于光谱直方图度量相邻分割块的相似性。最后融合标记信息和相似测度,以最小代价区域合并的方式提取目标。实验结果表明本文方法能够以很高的精度提取地灾目标,整体精度高于95%,Kappa系数优于0.90。
Abstract:
Geological disaster is usually extremely destructive, especially the landslip, landslide and mudslide.When it occurred, rapid and accurate extraction of the geological disaster information is not only pivotal for effective damage assessment but also critical for ensuring effective relief delivery.With the development of the aerospace technology, high-resolution remote sensing image has been widely accepted as a valuable source for disaster information extraction and damage assessment due to its merits of macro, quickness and high accuracy in disaster survey compared with traditional field survey.However, manual visual interpretation based on remote sensing images is time and labor consuming, while automatic computer interpretation is far from satisfying because of low precision.Concerning this issue, this paper proposed a semi-automatic geological disaster targets extraction algorithm.First, the geological disaster targets and background was manually marked on the remote sensing images with lines of different colors.Comparing with manual interpretation, manually marking is more labor-saving and high effective.Then mean shift algorithm was further applied to the remote sensing image to partition the whole image to multiple adjacent segments.Next, the similarity between the adjacent segments was measured with the utilization of spectral histogram.With the prior knowledge of the manual markers and the similarity measurement, the geological disaster targets were finally extracted by region merging procedure following the minimization criterion of merging cost.The proposed method was tested on various kinds of disasters with comparison of automatic object-oriented method and interactive grab cut method.The proposed semi-automatic approach not only took the advantage of the manual markers, which were important prior knowledge for targetidentification, but also benefits from the object-oriented image analysis paradigm.The experimental results showed that the proposed approach was capable of extracting the geological disaster targets with superiority of accuracy over automatic approach and grab cut algorithm.The overall accuracy was higher than 95% and Kappa coefficient is better than 0.9.

参考文献/References:

[1] 陶舒.汶川地震滑坡遥感信息提取及灾害危险性评价研究[M].北京:首都师范大学,2009:11-12[TAO Shu.Information extraction of landslides induced by the Wenchuan earthquake and susceptibility assessment using remotely sensed data [M].Beijing: Capital Normal University, 2009:11-12]
[2] 安立强, 张景发, 赵福军, 汶川地震次生灾害提取--面向对象影像分类技术的应用[J].自然灾害学报, 2011,20(02): 160-168 [AN Liqiang, ZHANG Jingfa, ZHAO Fujun.Extracting secondary disaster of Wenchuan earthquake:application of object-oriented image-classifying technology [J].Journal of Natural Disasters, 2011, 20(02): 160-168]
[3] 徐鹏杰,邓磊.遥感技术在减灾救灾中的应用[J].遥感技术与应用, 2011, 26(4): 512-519[XU Pengjie, DENG Lei.Application of remote sensing in the field of disaster reduction and response [J].Remote Sensing Technology and Application, 2011, 26(04): 512-519]
[4] 敖萌, 张勤, 赵超英,等.改进的CR-InSAR技术用于四川甲居滑坡形变监测[J].武汉大学学报(信息科学版), 2017,42(03):377-383[AO Meng, ZHANG Qin, ZHAO Chaoying,et al.An improved CR-InSAR technology used for deformation monitoring in Jiaju landslide, Sichuan[J].Geomatics and Information Science of Wuhan University, 2017,42(3): 377-383]
[5] 丁辉, 张茂省,李林.基于多特征面向对象区域滑坡现象识别[J].遥感技术与应用, 2013, 28(6): 1107-1113[DING Hui,ZHANG Maosheng, LI Lin.Regional landslide identification base on multi-feature object-oriented image classification[J].Remote Sensing Technology and Application, 2013, 28(6): 1107-1113]
[6] 王世博,张大明,罗斌,等.基于谱抠图的遥感图像滑坡半自动提取[J].计算机工程, 2012, 38(2): 195-197 [WANG Shibo, ZHANG Daming, LUO Bin, et al.Semi-automatic landslide extraction for remote sensing image based on spectral mattin [J].Computer Engineering, 2012, 38(02):195-197]
[7] 田力,张晓盼,袁艳斌.交互式遥感影像分割算法研究[J].武汉大学学报·信息科学版, 2014, 39(4): 406-410.[TIAN Li, ZHANG Xiaopan, YUAN Yanbin.Interactive remote sensing image segmentation algorithm[J].Geomatics And Information Science Of Wuhan Univers, 2014, 39(4): 406-410]
[8] RAU J Y, JHAN J P, RAU R J.Semiautomatic object-oriented landslide recognition scheme from multisensor optical imagery and DEM[J].IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(2): 1336-1349.
[9] ROTHER C, KOLMOGOROV V, BLAKE A.Grabcut: Interactive foreground extraction using iterated graph cuts[C].ACM transactions on graphics(TOG).ACM, 2004, 23(3): 309-314.
[10] LI Y, SUN J, TANG C K, et al.Lazy snapping[C].ACM Transactions on Graphics(ToG).ACM, 2004, 23(3): 303-308.
[11] Comaniciu D, Meer P.Mean shift: A robust approach toward feature space analysis[J].IEEE Transactions on Pattern Analysis And Machine Intelligence, 2002, 24(5): 603-619.
[12] Nummiaro K, Koller-Meier E, Van Gool L.An adaptive color-based particle filter[J].Image and vision computing, 2003, 21(1): 99-110.
[13] PENG B, ZHANG L, ZHANG D.Automatic image segmentation by dynamic region merging[J].IEEE Transactions on Image Processing, 2011, 20(12): 3592-3605.
[14] 张毅, 谭龙, 陈冠, 等.基于面向对象分类法的高分辨率遥感滑坡信息提取[J], 兰州大学学报(自然科学版), 2014, 50(05): 745-750[ZHANG Yi, TAN Long, CHEN Guan, et al.Landslide information extracted from high resolution remote sensing based on the object oriented classification method [J], Journal of LanZhou University(Natural Sciences), 2014, 50(05): 745-750]

相似文献/References:

[1]巩杰,赵彩霞,王合领,等.基于地质灾害的陇南山区生态风险评价——以陇南市武都区为例[J].山地学报,2012,(05):570.
　GONG Jie,ZHAO Caixia,WANG Heling,et al.Ecological Risk Assessment of Longnan Mountainous Area Based on Geological Disasters——A Case Study of Wudu[J].Mountain Research,2012,(04):570.
[2]牛全福,程维明,兰恒星,等.基于信息量模型的玉树地震次生地质灾害危险性评价[J].山地学报,2011,(02):243.
　NIU Quanfu,,et al.Susceptibility Assessment of Secondary Geological Disaster Based on Information Value Methodology for Yushu Earthquake Region[J].Mountain Research,2011,(04):243.
[3]孙伟,朱诚,王晓翠,等.南京幕府山地区地质灾害与地热调查[J].山地学报,2011,(06):753.
　SUN Wei,ZHU Cheng,MA Chunmei,et al.Investigation of Geothermal Resource and Geologic Hazards at Mufu Mountain Area in Nanjing[J].Mountain Research,2011,(04):753.
[4]许强,黄润秋.非线性科学理论在地质灾害评价预测中的应用——地质灾害系统分析原理[J].山地学报,2000,(03):272.
[5]许强,黄润秋,向喜琼.地质灾害发生时间和空间的预测预报[J].山地学报,2000,(S1):112.
[6]许强,黄润秋.地质灾害的非线性数据处理与建模技术[J].山地学报,2000,(S1):123.
[7]王帅,王深法,俞建强.构造活动与地质灾害的相关性——浙西南山地滑坡、崩塌、泥石流的分布规律[J].山地学报,2002,(01):47.
[8]柴贺军,刘汉超.岷江上游多级多期崩滑堵江事件初步研究[J].山地学报,2002,(05):616.
[9]邓辉,黄润秋.InSAR技术在地形测量和地质灾害研究中的应用[J].山地学报,2003,(03):373.
[10]马东涛,崔鹏,陈书涛,等.南水北调西线工程调水区地质灾害问题[J].山地学报,2003,(05):582.

备注/Memo

备注/Memo:
收稿日期(Received date):2017-03-08; 改回日期(Accepted date): 2018-8-26
基金项目(Foundation item):国家自然科学基金项目(41471353)。[National Natural Science Foundation of China(41471353)]
作者简介(Biography):王鹏(1990-),男,山东潍坊人,硕士,主要研究方向:遥感图像目标识别与信息提取。 [WANG Peng(1990-), male, born in Weifang, Shandong province, M.E., research on remote sensing image based target recognition and information extraction] E-mail: wangpeng187@126.com
更新日期/Last Update: 2018-07-30