| 基于多光谱遥感影像的巫峡滑坡灾害识别技术研究 |
| |
| 引用本文： | 郭健,张鹏,张全,黄波林,秦臻.基于多光谱遥感影像的巫峡滑坡灾害识别技术研究[J].华南地质与矿产,2020,36(1):38-45. |
| |
| 作者姓名： | 郭健 张鹏 张全 黄波林 秦臻 |
| |
| 作者单位： | 三峡大学土木建筑学院防灾减灾湖北省重点实验室,湖北宜昌443002,三峡大学土木建筑学院防灾减灾湖北省重点实验室,湖北宜昌443002,三峡大学土木建筑学院防灾减灾湖北省重点实验室,湖北宜昌443002,三峡大学土木建筑学院防灾减灾湖北省重点实验室,湖北宜昌443002,三峡大学土木建筑学院防灾减灾湖北省重点实验室,湖北宜昌443002 |
| |
| 基金项目： | 开放基金项目;国家重点研发计划项目 |
| |
| 摘 要： | 三峡库区地质结构复杂,尤其是巫峡高陡岸坡发育区域,历来是滑坡灾害高发区,对三峡库区的安全运行构成威胁。为了厘清巫峡高陡峡谷区滑坡灾害发育的高精度多光谱遥感影像特征,本研究以高精度遥感影像资料为基础,以巫峡高陡峡谷区为研究对象,应用面对对象的分类方法对研究区的高精遥感影像进行分割和分类,结合多尺度分割和ESP(Estimation of Scale Parameter)工具确定最优分割尺度,选取典型的滑坡对象样本进行最近邻分类,探索基于高精度多光谱遥感影像的高陡峡谷区滑坡灾害识别技术方法。经试验得出高陡峡谷区基于高精度多光谱遥感影像的滑坡灾害识别的最优分割尺度为720,形状因子和紧致度为0.5,在分割基础上,进行分类和滑坡信息提取。通过对比已有滑坡灾害资料,基于遥感影像自动识别滑坡灾害技术得到的研究区滑坡灾害分布结果总体精度达到了0.8696。研究结果表明,基于高精度的遥感影像自动识别滑坡灾害技术方法能够较好地进行分割与滑坡识别,精度评价的结果比较符合实际。研究结果为巫峡高陡峡谷区滑坡灾害的识别、调查、预测和防治提供依据,对三峡库区滑坡灾害的早期识别和防灾减灾有重要意义。
|
| 关 键 词： | 三峡库区 高陡峡谷区 巫峡 滑坡 遥感解译 |
Study on Landslide Hazard Identification Technology Based on Multispectral Remote Sensing Images in Wu Gorge |
| |
| GUO Jian,ZHANG Peng,ZHANG Quan,HUANG Bo-Lin,QIN Zhen.Study on Landslide Hazard Identification Technology Based on Multispectral Remote Sensing Images in Wu Gorge[J].Geology and Mineral Resources of South China,2020,36(1):38-45. |
| |
| Authors: | GUO Jian ZHANG Peng ZHANG Quan HUANG Bo-Lin QIN Zhen |
| |
| Institution: | (Hubei Key Laboratory of Disaster Prevention and Mitigation,College of Civil Engineering&Architecture,China Three Gorge University,Yichang 443002,Hubei,China) |
| |
| Abstract: | Because of its complicated geological structure,the Three Gorges Reservoir area has always been a region with high incidence of landslide hazards,especially the Wu Gorge high steep bank slope tract,which poses a threat to the safe operation of the Three Gorges Reservoir.In order to clarify the Wu Gorge based on high resolution multispectral remote sensing image the landslide development characteristics of high and steep canyon area,this study is based on high precision remote sensing image data,to Wu Gorge high and steep canyon area as the research object,apply to face object classification in the study area and high precision remote sensing image segmentation and classification,combined with multi-scale segmentation and ESP(Estimation of Scale Parameter)tool to determine the optimal segmentation scale,selection of typical landslide object nearest neighbor classification samples,thus set based on high resolution multispectral remote sensing image of high and steep canyon area landslide recognition method.According to the experiment,the optimal segmentation scale of landslide disaster identification based on high precision multi-spectral remote sensing image in high and steep canyon area is 720,shape factor and compactness are 0.5,so the distribution characteristics of landslide disaster in high and steep canyon area are obtained.On this basis,by comparing the existing landslide disaster data,the coincidence degree is better.At the same time,the accuracy of landslide disaster identification was evaluated in the study area.In general,the overall accuracy of landslide disaster distribution results in the study area based on the technology of automatic landslide disaster identification based on remote sensing image reached 0.8696.The results show that the method based on high precision remote sensing image can segment and identify landslide disaster better and faster.The research results provide a reliable basis for the identification,investigation,prediction and prevention of landslide hazards in the Wu Gorge high and steep canyon area,and are of great significance for the early identification and disaster prevention and reduction of landslide hazards in the Three Gorges Reservoir area. |
| |
| Keywords: | Three gorges reservoir area high and steep canyon area Wu gorge landslide remote sensing interpretation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
