| 基于地面激光扫描的典型海岸带盐沼潮滩地形反演 |
| |
| 引用本文： | 魏伟,周云轩,田波,钱伟伟,湛玉剑,黄盖先.基于地面激光扫描的典型海岸带盐沼潮滩地形反演[J].吉林大学学报(地球科学版),2018,48(6):1889-1897. |
| |
| 作者姓名： | 魏伟 周云轩 田波 钱伟伟 湛玉剑 黄盖先 |
| |
| 作者单位： | 华东师范大学河口海岸学国家重点实验室, 上海 200062 |
| |
| 基金项目： | 国家自然科学基金项目（41476151）；上海市科委科研计划项目（17DZ1201902，18DZ1204802） |
| |
| 摘 要： | 海岸带潮滩地形具有快速变化的特点；在盐沼植被覆盖的潮滩区域，地面激光扫描仪（terrestrial laser scanning，TLS）获取地形数据受到地表植被的影响。为了研究TLS在盐沼潮滩使用的地形测量精度与盐沼植被种类、盖度的关系，本文以芦苇群落、白茅群落、互花米草群落、海三棱藨草群落4种典型海岸带盐沼植被群落为研究对象，在移动窗口法的基础上辅以聚类分析的植被滤除算法，分别从点云原始数据中恢复地表地形地貌特征。研究结果表明：1）植被盖度越高，TLS反演地形精度越低，两者负相关。2）不同植被的激光穿透能力不同：盖度大于50%时，激光无法穿透原始盖度分别为70%、65%、65%的白茅群落、互花米草群落、海三棱藨草群落，均方根误差（root-mean-square error，RMSE）分别为22.0、22.0、8.6 cm；盖度等于50%时，白茅群落、海三棱藨草群落、芦苇群落和互花米草群落的RMSE分别为16.0、6.6、4.5、5.7 cm；盖度小于50%时，芦苇群落、互花米草群落、海三棱藨草群落地形反演精度小幅度提高，白茅群落地形反演精度提升较为明显。3）在盐沼潮滩地区使用TLS反演地形时，增加TLS架设高度、对同一区域多方位反复扫描可能有助于提高地形反演效果。
|
| 关 键 词： | 盐沼植被 地面激光扫描仪 植被滤除 地形 长江口海岸 |
| 收稿时间： | 2017-11-21 |
Topography Retrieval on Typical Salt Marsh of Coastal Zone Based on Terrestrial Laser Scanning |
| |
| Wei Wei,Zhou Yunxuan,Tian Bo,Qian Weiwei,Zhan Yujian,Huang Gaixian.Topography Retrieval on Typical Salt Marsh of Coastal Zone Based on Terrestrial Laser Scanning[J].Journal of Jilin Unviersity:Earth Science Edition,2018,48(6):1889-1897. |
| |
| Authors: | Wei Wei Zhou Yunxuan Tian Bo Qian Weiwei Zhan Yujian Huang Gaixian |
| |
| Institution: | State Key Laboratory of Estuary and Coastal Research, East China Normal University, Shanghai 200062, China |
| |
| Abstract: | The topography of tidal flat has the characteristics of rapid change. In the tidal flat area covered by salt marsh vegetation, acquiring topography data by terrestrial laser scanning (TLS) is affected by surface salt marsh. In order to study the precision of TLS on topography of salt marsh covered area and the relationship between its accuracy and vegetation coverage, we took the Phragmites australis, Imperata cylindrica, and Spartina alterniflora communities and Scirpus mariqueter as the research objects. The vegetation filter algorithm,based on the moving window method supplemented by cluster analysis, was used to recover the terrain and topography features from point cloud data. The results showed that:1) the higher the vegetation coverage,the lower the accuracy of TLS topography retrieval was, a negative correlation; 2) the penetration of TLS was different with the type of vegetation:when the vegetation coverage was greater than 50%, the original coverage that laser could not penetrate was 70%, 65% and 65% of Imperata cylindrica, Spartina alterniflora,and Scirpus mariqueter communities respectively, and the root-mean-square error (RMSE) was 22.0, 22.0 and 8.6 cm, respectively; when coverage equaled to 50%, RMSE of topography retrieval was 16.0 cm of Imperata cylindrica community, 6.6 cm of Scirpus mariqueter community, 4.5 and 5.7 cm respectively of Phragmites australis and Spartina alterniflora communities; when coverage was less than 50%, with the decrease of the coverage, the accuracy of topography retrieval of the Phragmites australis, Spartina alterniflora and Scirpus mariqueter communities increased slightly, while that of the Imperata cylindrica community was improved more obviously. 3) When using TLS recover topography in the salt marsh area, the higher and the more of the TLS setting direction in the same area, the better the accuracy of the topography retrieval was. |
| |
| Keywords: | salt marsh vegetation terrestrial laser scanning vegetation filtration topography Yangtze River estuary coast |
|
| 点击此处可从《吉林大学学报(地球科学版)》浏览原始摘要信息 |
| 点击此处可从《吉林大学学报(地球科学版)》下载免费的PDF全文 |
|
