无人机遥感数据处理与滑坡信息提取

发布时间：2018-04-10 19:02

本文选题：无人机遥感 + 滑坡信息提取　；参考：《地球信息科学学报》2017年05期

【摘要】：高分辨率的DEM和DOM数据是对地形地貌信息的准确描述,也是滑坡信息提取的重要数据源。首先,针对滑坡信息提取的要求,本文采用无人搭载微型单反相机的影像获取平台,结合野外测量的GPS数据,弥补了无人机POS信息精度低的劣势;针对无人机影像的特点,运用摄影测量基本原理与计算机视觉算法,获取高精度、高分辨率的DEM与DOM影像,保留了丰富的光谱与纹理信息。其次,借助ESP辅助工具获取了DOM影像的最佳分割尺度,并结合研究区地物特征构建了基于模糊分类与SVM算法相结合的决策树,运用面向对象的分类方法实现了对研究区内植被、道路、疑似滑坡区域的信息提取。最后,依照研究区地物分布的空间特征确定了高风险等级区域,并对该区域进行滑坡的形态与纹理分析以及精度评价,其中提取的疑似滑坡区域用户精度为91.44%、生产者精度为84.65%,结果表明无人机遥感在滑坡信息提取领域具有较高的应用价值。
[Abstract]:High resolution DEM and DOM data are not only the accurate description of landform information, but also an important data source for landslide information extraction.First of all, according to the requirements of landslide information extraction, this paper adopts the image acquisition platform of unmanned micro SLR camera, combined with field measurement of GPS data, to make up the disadvantage of low precision of UAV POS information, aiming at the characteristics of UAV image.The basic principles of photogrammetry and computer vision algorithm are used to obtain DEM and DOM images with high accuracy and high resolution, and abundant spectral and texture information are retained.Road, suspected landslide area information extraction.Finally, according to the spatial characteristics of the distribution of ground objects in the study area, the high risk grade area is determined, and the shape and texture analysis and accuracy evaluation of the landslide are carried out.The user accuracy of suspected landslide area is 91.44 and the precision of producer is 84.65. The result shows that UAV remote sensing has high application value in landslide information extraction field.
【作者单位】：首都师范大学资源环境与旅游学院;首都师范大学
【基金】：国家科技支撑计划项目(2013BAC03B04) 国家自然科学基金项目(41301468)
【分类号】：P237;P642.22

【相似文献】