以無人機應用崩塌土方量推估與後期植生復原監測

臺灣地形陡峭，又因九二一大地震使全臺地層鬆動，導致每逢極端降雨時，常產生崩塌災害。為即時掌握災後資訊，本文利用具機動性、高解析力之無人機（Unmanned Aerial Vehicle, UAV），於2017年6月拍攝南投縣鹿谷鄉小半天之災後崩塌情形。經分析，崩塌總面積為20,983 m^2，其中7,007 m^2的面積為裸露崩塌區域，陷落區面積為13,976 m^2，導致崩塌上游形成堰塞湖，其面積 1,114 m^2。經UAV拍攝現場產製的數位地表模型（Digital Surface Model, DSM）資料，與災害前之數值高程模型（Digital Elevation Model, DEM）作土方量分析，崩塌土方遞移量為17,032 m^3。經UAV即時掌握災後資訊，即可迅速進行決策，施作因應工程及架設自動監控系統，若未來發生災害，即可透過自動監控系統，並搭配防災專員協助疏散居民。

關鍵字

無人機；數位地表模型；土方量；崩塌地；植生復原監測

並列摘要

Taiwan often suffers from landslides after extreme precipitation due to the widespread steep terrain and stratum loosening caused by the 921 earthquake. To collect instant disaster information, this study utilized highly mobile UAVs with high camera resolution to record the landslide that occurred in Hsiao Pan Tien, Lugu Township, Nantou County in June 2017. The analysis revealed that the landslide area was 7,007 m^2 , with a 13,976 m^2 zone of depletion at the crown and the formation of a 1,114 m^2 landslide dam. By comparing the Digital Surface Model (DSM) captured by the UAV with the Digital Elevation Model (DEM) made before the disaster, the analysis showed that the landslide volume was 17,032 cubic meters. The UAV proved to be capable of collecting real-time post-disaster information, which can assist in rapid decision-making. By implementing onsite devices and setting up automatic monitoring systems, disaster prevention specialists can use this information for resident evacuation.

並列關鍵字

unmanned aerial vehicle ； digital surface model ； landslide volume ； landslide ； vegetation recovery monitoring

參考文獻

施虹如、吳亭燁、蘇元風、劉哲欣、李欣輯、陳永明、張志新，2015，〈極端降雨事件下淺層崩塌潛勢衝擊評估分析〉，《工程環境會刊》，34，頁 77-95。doi:10.6562/JEE.2015.34.5

Hsiao C.-Y., Lin B.-S., Chen C.-K., and Chang D.-W., 2014, “Application of Airborne LiDAR Technology in Analyzing Sediment-Related Disasters and Effectiveness of Conservation Management in Shihmen Watershed,” Journal of GeoEngineering, 9(2), 55-73. doi:10.6310/jog.2014.9(2).2

Heimonen T., Hannuksela J., Heikkilä J., Leinonen J., and Manninen M., 2001, “Experiments in 3D Measurements by Using Single Camera and Accurate Motion,” in Proceedings of the 2001 IEEE International Symposium on Assembly and Task Planning, Piscataway, NJ: IEEE, 356-361. doi:10.1109/ISATP.2001.929051

Szeliski R., 2010, Computer Vision: Algorithms and Applications, London, UK: Springer-Verlag London. doi:10.1007/978-1-84882-935-0

DJI，n.d.，〈Phantom 4 Pro V2.0〉，《DJI》，https://www.dji.com/tw/phantom-4-pro-v2?site=brandsite&from=landing_page（瀏覽日期：2023 年 1 月 30 日）。

Google Scholar

主題瀏覽